Aspects of Multinational Enterprises in the Global …etheses.lse.ac.uk/3266/1/Ascani_Aspects_of_Multinational.pdf3 Abstract The role played by Multinational Enterprises (MNEs) in

London School of Economics and Political

Science

Aspects of Multinational Enterprises in the Global

Economy:

Location, Organisation and Impact

Andrea Ascani

A thesis submitted to the Department of Geography & Environment of the London School of Economics for the degree of Doctor of Philosophy, London, July 2015.

2

Declaration

I certify that the thesis I have presented for examination for the MPhil/PhD

degree of the London School of Economics and Political Science is solely my own

work other than where I have clearly indicated that it is the work of others (in

which case the extent of any work carried out jointly by me and any other person

is clearly identified in it).

The copyright of this thesis rests with the author. Quotation from it is

permitted, provided that full acknowledgement is made. This thesis may not be

reproduced without my prior written consent.

I warrant that this authorisation does not, to the best of my belief, infringe the

rights of any third party.

I declare that my thesis consists of 51,901 words.

Statement of conjoint work

I confirm that Chapters 1, 2 and 3 were jointly co-authored with Professor

Iammarino and Dr Crescenzi. I contributed 50% of the work in Chapters 1 and 2,

and 66% in Chapter 3.

I also confirm that Chapter 5 was jointly co-authored with Dr Gagliardi and I

contributed to 50% of this work.

London, 15th July 2015

Andrea Ascani

3

Abstract

The role played by Multinational Enterprises (MNEs) in the global

economy is becoming increasingly relevant as they shape sectorial,

regional and national trajectories of economic development through their

cross-border activities and behaviour. This thesis investigates how the

characteristics of MNEs, their activities and location-specific attributes

interact with each other and shape both behaviour and choices of MNEs

and the impact of foreign direct investment (FDI). The thesis is

structured into a general Introduction, aimed at guiding the reader

throughout the thesis and providing a broad conceptual framework, and

three analytical Parts focusing on (i) MNE greenfield investment location

strategies, (ii) MNE selection decisions in cross-border acquisitions and

(iii) impact of MNE operations on host regions.

In Part I, the location behaviour of MNEs, in the light of the

specificities of the recipient economies, is carefully analysed. In

particular, the three Chapters of Part I investigate the location behaviour

of European MNEs in a set of European Union (EU) neighbouring

countries over the period 2003-2008, by focusing on different aspects of

location strategies. In Chapter 1, an initial descriptive analysis is

produced in order to account for the general determinants of MNE

location behaviour. This chapter, therefore, offers a quantitative

assessment of the main drivers of FDI in the EU neighbourhood and it

also explores sectorial and functional dynamics. Chapter 2 deepens the

study of MNE location behaviour by developing both a quantitative and a

qualitative analysis of FDI determinants based on the experience of

Italian MNEs operating in the EU neighbourhood. This mixed-methods

approach allows integrating the general insights emerging from the

analysis of the broad group of Italian investors with the in-depth case

studies of two specific large Italian MNEs with a strong presence in EU

4

neighbouring countries in recent years. Subsequently, in Chapter 3,

particular attention is devoted to the empirical analysis of the spatial

distribution of MNE activities in relation to differences in terms of

economic institutions of the host locations. This specific line of research

is based on an innovative quantitative approach to the study of MNE

location strategies in terms of greenfield FDI in the sample of

neighbouring countries of the EU. In particular, Chapter 3 focuses on the

heterogeneous location strategies of MNEs with respect to location

attributes. Overall, the main findings of Part I of the thesis not only

suggest that the traditional drivers of FDI emphasised in the existing

literature, such as market access and cost-saving factors, still represent

relevant elements for MNE behaviour, but it is also highlighted that MNE

specificities are crucial to understand investment choices and that

industry-wide differences can influence both entry modes and the

location decisions of MNEs. The most innovative contribution of Part I,

however, is related to Chapter 3, where the quantitative analysis of MNE

location behaviour by means of Mixed Logit models suggests that MNEs

have heterogeneous preferences with respect to location characteristics,

especially economic institutions. This indicates that MNE strategies are

highly diverse and the previous quantitative literature may have

underestimated the complexity of the interaction between MNEs

characteristics and location attributes.

After exploring the determinants of MNE location strategies, Part II of

the thesis aims at studying the selection decisions of MNEs engaging in

cross-border acquisitions. This represents a very novel area of enquiry

and the objective of Chapter 4 is to quantitatively assess the relevance of

target firms’ attributes in shaping MNE acquisition choices in the

framework of their international organisation of production. In

particular, the aim of this Chapter is to assess whether acquisition

decisions are associated to the search of strategic assets or to market

access considerations. Results suggest that, in the sample of EU15 firms

5

under analysis in the period 1997-2013, the latter motivation tends to be

more relevant. This is in line with market access motives operating at the

firm level, differently from other studies on FDI and acquisitions focusing

on the industry- or country-wide level of analysis. Evidence in favour of

strategic-asset seeking strategies of MNEs acquiring European firms,

instead, remains weak. Therefore, this Chapter highlights that domestic

firms engaging in the generation of successful business linkages within

or across national markets can represent a valuable target for MNE

cross-border acquisition decisions.

Finally, building on the previous sections on the determinants of

location choices and selection patterns in cross-border takeovers, Part III

of the thesis focuses on the impact of FDI on recipient areas in terms of

their innovation potential. Chapter 5 is developed as a quantitative

analysis with the specific objective of isolating the causal effect of MNE

operations on the innovative performance of host regions. This is

investigated by employing NUTS-3 level data on Italy for the period 2001-

2006. The empirical analysis is supported by the implementation of an

Instrumental Variable (IV) strategy in order to tackle potential

endogeneity bias in the estimation of FDI-induced spillovers. This

Chapter contributes to the existing debate by focusing on the

geographical level of FDI externalities, whereas the great majority of past

studies mainly investigate industry-wide effects. Results suggest that the

presence of FDI in a location contributes to fostering the innovative

performance of the local economy. Therefore, MNEs can be seen as

carriers of superior knowledge and new organisational practices that spill

over space to the benefit of domestic firms. In a policy-making

perspective, this provides a clear rationale for the attraction of FDI as an

international channel for knowledge sourcing.

The three Parts of the thesis are strongly complementary as the

strategies of MNEs in Part I and II in terms of FDI (i.e. greenfield and

acquisitions) are integrated with an assessment of the impact that

6

corporate activities have on recipient economies in Part III. Although the

broad conceptual background to the work as a whole is provided in the

general Introduction of the thesis, each Chapter has a section devoted to

a dedicated and specific review of the literature. Moreover, the thesis also

contains an acknowledgement of the limitations of the study, which is

provided in the concluding sections of each Chapter, as well as a

discussion of the contributions and implications that the analyses

developed in the various Chapters have for academic research and

policy-making.

7

Contents

Contents......................................................................................... 7

List of Tables ................................................................................ 10

List of figures ............................................................................... 12

Acknowledgements ....................................................................... 13

Introduction ................................................................................. 15

I. Overview ....................................................................................... 15

II. Broad conceptual framework ........................................................ 16

III. Aim and structure of the thesis ................................................... 20

Part I - MNE location strategies ....................................................... 21

Part II – Selection patterns in cross-border acquisitions .................... 23

Part III – The impact of FDI on recipient economies .......................... 23

IV. Concluding remarks .................................................................... 24

Part I: MNE Location Strategies .................................................... 27

Chapter 1 - The geography of foreign investments in the EU Neighbourhood ............................................................................. 28

1.1 Introduction ............................................................................. 28

1.2 Literature background: the drivers of FDI into the EU

neighbourhood ................................................................................. 30

1.3 Stylised facts on FDI in the EU neighbourhood ........................ 33

1.4 FDI in the EU neighbourhood: methodology ............................. 36

1.5 Results .................................................................................... 39

1.6 Conclusions ............................................................................. 44

Appendix A....................................................................................... 53

Chapter 2 – What drives European multinationals to the EU neighbouring countries? A mixed methods analysis of Italian

investment strategies ................................................................... 54

2.1 Introduction ............................................................................. 54

2.2 Italian Foreign Investments in EU New Member States and Neighbouring Countries ................................................................... 58

2.3 Quantitative analysis ............................................................... 62

2.3.1 Empirical model and data..................................................... 62

8

2.4 Results and discussion ............................................................ 66

2.5 Qualitative analysis ................................................................. 69

2.5.1 MNEs profiles ....................................................................... 69

2.5.2 Analysis of the interviews with executives ............................. 72

2.6 Conclusions ............................................................................. 78

Appendix B ...................................................................................... 86

Chapter 3 – Economic Institutions and the Location Strategies of

European Multinationals in their Geographical Neighbourhood ..... 88

3.1 Introduction ............................................................................... 88

3.2 MNEs location strategies, host economy advantages and institutional conditions .................................................................... 91

3.2.1 MNEs and host economy advantages .................................... 91

3.2.2 Economic institutions and MNEs investments ...................... 94

3.3 Data ........................................................................................ 99

3.3.1 MNE Investment ................................................................... 99

3.3.2 Institutional Conditions .......................................................101

3.3.3 Other location drivers ..........................................................103

3.4 Methodology ...........................................................................106

3.4.1 Capturing MNEs heterogeneous preferences for economic institutions: Mixed Logit Models ...................................................106

3.5 Empirical Results ...................................................................109

3.5.1 Baseline results ...................................................................109

3.5.2 Preference heterogeneity ......................................................111

3.6 Conclusions ............................................................................118

Appendix C .....................................................................................131

Part II: Selection Patterns in Cross-border Acquisitions ...............134

Chapter 4 – Cross-border acquisitions and patterns of selection:

Productivity vs. profitability ........................................................135

4.1 Introduction ............................................................................135

4.2 Related literature.......................................................................138

4.2.1 Acquisitions to access foreign productive assets ..................139

4.2.2 Acquisitions to access foreign markets ................................142

4.2.3 Hypotheses development .....................................................143

4.3 Data ..........................................................................................144

4.3.1 Dataset construction ...........................................................144

9

4.3.2 Variables construction ........................................................146

4.4 Empirical strategy .....................................................................149

4.5 Results ......................................................................................152

4.5.1 Probability of foreign acquisition: baseline estimates ...........152

4.5.2 Evidence from alternative measures of profitability and

productivity..................................................................................155

4.5.3 Non-linearity in within-firm probability of foreign acquisition ....................................................................................................158

4.5.4 Foreign acquisitions and technology ....................................160

4.5.5 Completed and majority foreign acquisitions .......................162

4.5.6 Evidence from acquisition targets only.................................163

4.6 Conclusions ..............................................................................164

Part III: The Impact of FDI on Recipient Economies ....................178

Chapter 5 – Inward FDI and Local Innovative Performance. An

empirical investigation on Italian provinces ................................179

5.1 Introduction ............................................................................179

5.2 Conceptual background and literature review .........................181

5.3 Data .......................................................................................184

5.4 Methodology ...........................................................................187

5.5 Results and discussion ...........................................................191

5.6 Robustness Checks .................................................................195

5.7 Conclusion .............................................................................197

Bibliography ................................................................................208

10

List of Tables

Table 1.1: FDI into the EU neighbourhood ........................................... 47

Table 1.2: FDI into the EU neighbourhood by business function .......... 48

Table 1.3: FDI into the EU neighbourhood by macro-sector ................. 49

Table 1.4: FDI determinants into the EU neighbourhood ..................... 50

Table 1.5: FDI determinants in the EU neighbourhood by business

function .............................................................................................. 51

Table 1.6: FDI determinants in the EU neighbourhood by macro-sector 52

Table 2.1: Italian new foreign operations in the EU NMs and NCs ........ 81

Table 2.2: Italian new foreign operations in the EU NMs and NCs by

business activity.................................................................................. 82

Table 2.3: Italian new foreign operations in the EU NMS and NCs by

sector .................................................................................................. 83

Table 2.4: Poisson regression results ................................................... 84

Table 2.5: Summary Table of Case Studies .......................................... 85

Table 3.1: EU-15 investment projects and quality of economic

institutions, 2003-2008. .....................................................................122

Table 3.2: Conditional Logit estimation of EU15 MNEs location behaviour

..........................................................................................................123

Table 3.3: Mixed Logit estimation of MNEs location behaviour ............124

Table 3.4: MXL estimation of EU-15 MNEs location behaviour by sector

..........................................................................................................126

Table 3.5: MXL estimation of EU-15 MNEs location behaviour by

business function ...............................................................................128

Table 3.6: Summary Table of the Results on MNEs heterogeneous

preferences for Economic Institutions .................................................130

Table 4.1: Firms and acquisitions by country, 1997-2013 ...................168

Table 4.2: Correlation between measures of profitability and labour

productivity ........................................................................................169

11

Table 4.3: Descriptive statistics ..........................................................170

Table 4.4: Probability of foreign acquisition.........................................171

Table 4.5: Alternative measures for profitability and labour productivity

..........................................................................................................172

Table 4.6: Foreign acquisitions and total factor productivity ...............173

Table 4.7: Interaction effect between firm profitability and labour

productivity ........................................................................................174

Table 4.8: Probability of foreign acquisition by technological class ......175

Table 4.9: Completed and majority acquisitions ..................................176

Table 4.10: Restricted sample .............................................................177

Table 5.1: Variables List .....................................................................202

Table 5.2: Inward FDI and Local Innovative Performance ....................203

Table 5.3: First Stage Regression ........................................................204

Table 5.4: First Stage Statistics ..........................................................204

Table 5.5: Model Specification ............................................................205

Table 5.6: Reduced Form Equation .....................................................206

Table 5.7: Market Exit ........................................................................206

Table 5.8: Inward FDI and Labour Productivity ...................................207

12

List of figures

Figure 1: Growth of FDI, trade and GDP in the world, 1970-2010 ........ 26

Figure 3.1: Probability Density Functions for economic institutions

exhibiting significant standard deviation in Table 3 ............................125

Figure 3.2: Probability Density Functions for economic institutions

exhibiting significant standard deviation in Table 4 ............................127

Figure 3.3: Probability Density Functions for economic institutions

exhibiting significant standard deviation in Table 5 ............................129

Figure 5.1: Share of Inward FDI per Macro-Region ..............................201

13

Acknowledgements

I would like to thank to Simona Iammarino and Riccardo Crescenzi

for outstanding supervision and encouragement, and all members of my

family for unconditional love and tireless support.

Thanks to Giulia Faggio, Luisa Gagliardi, Steve Gibbons, Neil Lee,

Henry Overman, Olmo Silva and Michael Storper for help, advice and

ideas along the way.

I also would like to thank Davide Castellani and Xiaming Liu for their

feedback and suggestions.

Thanks also to Chinchih Chen, Marco Di Cataldo, Mara Giua, Alex

Jaax, Davide Luca, Paolo Lucchino, Paula Prenzel, Alessandra Scandura,

Teresa Schlueter, Alessandro Sforza, Maria Sanchez-Vidal and Stefanie

Vollmer for friendship and support.

Previous versions of the chapters of this thesis were presented at the

following conferences and seminars: 2nd SEARCH meeting, University of

Cagliari & CRENoS (September, 2012); Workshop on Innovation,

Productivity and Growth in Italy, University of Calabria (March 2013); XIV

April International Academic Conference on Economic and Social

Development, Moscow Higher School of Economics (April, 2013); XI Triple

Helix Conference, Birkbeck College, London (July, 2013); 53rd ERSA

Congress, University of Palermo (August, 2013); IV EUGEO Congress,

14

University of Rome ‘La Sapienza’ (September, 2013); Regional Studies

Association Winter Conference, London (November, 2014); Academy of

International Business Conference, Bangalore (June, 2015); 62nd Annual

North American Meeting of the Regional Science Association

International; various PhD Work-in-Progress seminars in Economic

Geography, London School of Economics. Thanks to participants and

discussants for helpful feedback.

The research leading to the results of the first three chapters has

received funding from the Project “Sharing KnowledgE Assets:

InteRregionally Cohesive NeigHborhoods” (SEARCH) within the 7th

European Community Framework Programme FP7-SSH - 2010.2.2-1

(266834) European Commission.

This PhD thesis was also supported by funding from the doctoral

studentship of the UK Economic and Social Research Council (ESRC).

All errors and omissions remain my own.

15

Introduction

I. Overview

This thesis explores how Multinational Enterprises (MNEs) shape the

international organisation of economic activity through Foreign Direct

Investment (FDI), by focusing on a number of relevant aspects that are –

to different extents – still partially addressed by existing studies, or

subject to mixed and inconclusive empirical evidence. This thesis aims at

filling some of the research gaps that characterise the literature, albeit

this being vast and well-established. The thesis’ structure consists of the

present introductory section and five empirical chapters divided into

three conceptual parts associated with different aspects of MNE activity.

Each chapter of the thesis includes an introduction, a scrutiny of the

literature with a presentation of the hypotheses, a description of data

and methodology, a discussion of results and a final section devoted to

concluding remarks, limitations and future research directions.

The relevance of MNEs in the global economy has dramatically

increased in the last decades, as evidenced by the astonishing spur in

the global growth rate of FDI since the mid-1980s and the consequent

outpacing of world exports and nominal GDP growth rates. Figure 1

illustrates this noticeable trend employing data from the United Nations

Conference on Trade and Development (UNCTAD), the well-known

international organisation that, in response to the unprecedented role

played by MNEs in the world economy, inaugurated in 1991 a series of

yearly studies to debate the characteristics, drivers and trends of FDI,

and currently publishing the 25th edition of the World Investment Report.

For the purpose of this thesis an MNE is intended in its simplest

definition as a firm that engages in activities across national borders

through FDI. In this respect, the firm undertaking FDI is the parent

16

company located in the country of origin, while the firm receiving the FDI

is defined as foreign affiliate or subsidiary and it is located in the

destination country1. At the simplest level, FDI modes can be classified

into greenfield investment and cross-border mergers and acquisitions

(M&A). The former encompasses the establishment a completely new

plant in a foreign location, whereas the latter entails the acquisition of a

certain stake of ownership in a pre-existing company abroad.

The objective of this introductory section is to provide a general

background framework for the thesis, describe its motivation, explain the

research aims and illustrate the structure and main content of the

various chapters. In particular, the next section discusses the basic

ideas that underpin the conceptualisation of MNEs in academic research.

Subsequently, the structure of the thesis is described and a summary of

each chapter’s objective, results and original contribution is offered.

Finally, a concluding section summarises the inner logic of the thesis, its

novelty and outlines some directions for future research.

[Figure 1 here]

II. Broad conceptual framework

The existence and importance of MNEs has received the attention of

scholars for decades, even before the enormous global increase in

multinational activity. A plethora of conceptual explanations, drawing on

diverse theoretical traditions, has been provided over the years to

understand and analyse the behaviour and strategies of MNEs. A

fundamental theoretical and empirical puzzle that academic research has

attempted to solve is associated to the existence of firms that decide to

become multinational. The tentative explanations of this aspect have

1 For the purpose of this thesis we use the notions of foreign affiliate and foreign subsidiary

interchangeably.

17

been underpinned by numerous hypotheses formulated over time. The

aim of this section is to review the main conceptual contributions to this

debate in order to provide a general framework for the various chapters

of the thesis, where more detailed aspects of MNEs will be analysed.

Therefore, this section will clarify the conceptual factors that have been

hypothesised by scholars as crucial for MNEs to exist, while the specific

conceptual frameworks associated to the distinct aspects investigated in

this thesis are developed in dedicated sections within the various

chapters.

The seminal work of Hymer (1976/1960) and Kindleberger (1969)

provides the starting point for a conceptualisation that explains

consistently why some firms engage in cross-border activities. Their basic

insight is that domestic firms tend to have specific advantages over

foreign firms when serving their domestic market. These advantages are

embodied in the domestic nature of local firms and range from better

information about the local economy and customers’ tastes to greater

familiarity with the political and legal system. Hence, foreign firms that

wish to operate in foreign markets have to offset their disadvantages over

domestic actors by increasing their efficiency. This is possible through

the acquisition of firm-specific advantages, which may vary from

economies of scale and product differentiation, to technological

advantages and access to cheaper factors of production. While insightful,

this conceptualisation does not help to explain why firms decide to locate

in a foreign country. In fact, even if foreign firms have specific

advantages over domestic firms, they may prefer to serve distant markets

by exports.

Another seminal contribution to the economic theory of MNEs is the

product life-cycle model of Vernon (1966, 1979). He considers three main

stages of a product life-cycle. First, when a product is new it is mostly

produced and sold by the most innovative firms in the home country

18

(normally a developed country). In the second stage, the product becomes

mature and it is exported. In this phase, demand grows in foreign

markets and the firm may decide to invest abroad to serve those markets

locally: thus, in this stage production gradually moves to foreign

countries (mainly other developed economies). Third, the product is

standardised and more firms are able to produce it. As a consequence,

price competition leads firms to invest in locations that make a reduction

in production costs possible (mainly developing countries). While this

theory provides an insightful conceptualisation of MNEs in innovative

industries, it does not offer a strong explanation for FDI in lower

technology sectors. Furthermore, this theory entails a simplistic and

reductive view of the innovation process, overlooking the complexity of

MNE innovative activities (Iammarino and McCann, 2013)

The occurrence of FDI has also been explored in terms of attempts of

firms to limit the market power of their competitors. According to this

hypothesis, oligopolistic firms follow similar FDI strategies as a way to

countering the advantages of other competing firms. Therefore, foreign

investment is considered as an oligopolistic reaction with the aim of

offsetting the competitive edge of similar firms (e.g. Knickerbrocker,

1973; Flowers 1976; Yu and Ito, 1988). An important limitation of this

theory is that its logic implies that more intense competition on world

markets is very likely to lead to less oligopolistic reaction and, as a

consequence, lower volumes of FDI. However, direct observation of world

trends shows that nowadays there is stronger competition and higher

volumes of FDI.

A highly relevant contribution to the explanation of why firms become

multinationals is provided by the hypothesis of internalisation of external

markets (Buckley and Casson, 1976; Casson, 1979; Rugman, 1981).

Fundamentally, the existence of imperfect markets implies higher costs

to link activities and exchanges across geographically separate markets.

Hence, firms decide to internalise these markets within their

19

organisational structure and to operate exchanges within the boundaries

of the firm across national borders. In other words, firms become

multinationals in order to avoid imperfections such as market

uncertainty, wastes of time and resources and asymmetric information.

In this sense, some firms prefer to open a subsidiary in another country

and to trade with it rather than licensing to local firms or exporting.

Underpinned by the insight of Hymer and Kindleberger on firm-

specific advantages and the idea of internalisation of external markets,

Dunning (1977, 1980 and 1988) elaborate the most widely accepted and

comprehensive economic framework of the origin of MNEs. His well-

known OLI eclectic paradigm entails that firms must satisfy three

conditions to become multinationals: (i) they have to possess owner-

specific advantages (O), (ii) some location-specific advantages should be

available (L) and (iii) they have to find profitable to internalise the use of

ownership advantages (I). This seminal conceptualisation made by

Dunning still provides a coherent and well-established answer to the

issue of the existence of MNEs. The existence of ownership-specific

advantages (O) possessed by some firms may lead to the decision to

internalise (I) these advantages and to locate in foreign markets as a way

to maximize their productive efficiency and to limit the impact of

uncertain and imperfect markets on production. In other words, FDI

occurs when firms possess assets of their own, and consider as more

convenient to internalise the use of such advantages rather than selling

or sub-contracting them to external companies. At the same time, these

firms decide to locate abroad where location-specific factors (L) allow for

a more profitable utilisation of the afore-mentioned ownership

advantages. In this perspective the (O), the (L) and the (I) are all

fundamental conceptual categories to explain the existence of MNEs and

the reasons why they undertake foreign investments. As a matter of fact,

according to Dunning himself “the OLI triad of variables […] may be

likened to three-legged stool: each leg is supportive of the other, and the

20

stool is only functional if the three legs are evenly balanced” (Dunning,

2009:5). The eclectic OLI paradigm, therefore, provides a convincing and

flexible conceptualisation of MNE existence and behaviour, although

being lacking under other aspects. For instance, the geography of MNEs

remains loosely specified in its (L) advantages, calling for further for

investigation (Iammarino and McCann, 2013).

More recently, the study of MNE has also grown in the international

trade literature, where the combination of the Krugman (1980) model

based on product differentiation and monopolistic competition with the

notion of firm heterogeneity (Melitz, 2003) has allowed to overcoming

formal problems in modelling MNE activity. In this respect, a relevant

implication of firm heterogeneity for the study of MNE is related to the

intra-industry diversity of internationalisation modes as a response to

differences in the accumulation of knowledge across MNEs (Castellani

and Zanfei, 2006).

III. Aim and structure of the thesis

While the academic literature studying the operations of MNEs is

large, this thesis identifies a number of research gaps associated with

specific aspects of multinational activity. The specific contribution that

the thesis will offer to the academic debate is discussed in each of the

chapters that constitute the main body of this work. Nevertheless, in

explaining the general structure and aims of the thesis, this section will

briefly discuss the main points of novelty developed in the various

chapters. In general, this work contributes to the literature on MNEs and

FDI and, particularly, on the different streams of research that mainly

contribute to this topic, such as economic geography, international

economics and international business and management studies.

21

As mentioned above, the thesis is divided into three main parts

containing five chapters. The first part contains three chapters while the

remaining two parts are constituted by one chapter each.

Part I - MNE location strategies

In the first part, this thesis examines the location behaviour of

European MNEs with respect to a number of drivers that are under-

explored in the literature. The first chapter offers an explorative analysis

of MNE location choices in countries linked to the ‘core’ of the European

Union (EU-15) by different degrees of functional, economic and political

integration: the EU 'New' Member states, Accession and Candidate

countries, European Neighbourhood Policy countries, as well as Russia.

Understanding the drivers of Foreign Investment (FDI) in these countries

is highly relevant in consideration of their increasing integration into the

global market and the strong influence exerted by the EU on this

process. By employing data on individual greenfield investment projects,

this chapter aims at disentangling the drivers of FDI in these countries

for different industrial sectors, business functions and investment

origins. The empirical results suggest that FDI in the area tends to follow

market-seeking and efficiency-oriented strategies, and show path-

dependency and concentration patterns that may reinforce core-

periphery development trajectories in the EU neighbourhood.

The second chapter narrows the analysis down to a specific case

study of an ‘old’ EU member country, Italy, investing in the same

destination area analysed in the first chapter. In so doing, this second

chapter adopts a mixed methods strategy combining a descriptive

statistical analysis with interviews with selected MNEs. Thus, the

analysis investigates the economic integration between Italy and the EU

neighbouring countries by exploring the location drivers of Italian-owned

MNEs in 33 destination economies including the New Member States of

22

the EU and the European Neighbouring countries. The paper compares

market-seeking and efficiency-seeking motivations with asset-seeking

strategies. The quantitative analysis assess the location determinants of

518 Italian MNEs that invested in the area in the 2003-2008 period,

while qualitative information on strategic location decisions is collected

by means of in-depth interviews with executives in two of the largest

Italian MNEs active in the region. The evidence suggests that market-

seeking considerations are still predominant drivers of Italian MNE

location decisions in EU Neighbouring Countries, together with resource-

seeking motivations. However, different MNEs are developing diversified

strategies to increase their access to these areas which are of increasing

interest for global investors.

The third chapter offers the most structured analysis of MNE location

behaviour looking at a neglected factor in the literature. This chapter, in

fact, examines how the location behaviour of MNEs is shaped by the

economic institutions of the host countries. The analysis still covers a

wide set of geographically proximate economies with different degrees of

integration with the ‘Old’ 15 European Union members: New Member

States, Accession and Candidate Countries, as well as European

Neighbourhood Policy countries and the Russian Federation. The

analysis aims at shedding light on the heterogeneity of MNE preferences

for the host countries’ regulatory settings (including labour market and

business regulation), legal aspects (i.e. protection of property rights and

contract enforcement) and the extent of government intervention in the

economy. By employing data on 6,888 greenfield investment projects, the

random-coefficient Mixed Logit analysis here applied shows that, while

the quality of the national institutional framework is generally beneficial

for the attraction of foreign investment, MNEs preferences over economic

institutions are highly heterogeneous across sectors and business

functions.

23

Part II – Selection patterns in cross-border acquisitions

After exploring the determinants of MNEs location decisions, this

chapter addresses the patterns of selection of cross-border acquisition

operations undertaken by MNEs. This represents a very novel area of

enquiry and the objective of this chapter is to quantitatively assess the

relevance of target firms’ attributes in shaping the acquisition choices of

MNEs in the framework of their international organisation of production.

By employing firm-level data on EU-15 countries, this fourth chapter

studies the extent to which different firm-level attributes of domestic

target companies motivate cross-border takeovers. In so doing, this work

analyses changes in ownership from domestic to foreign in a sample of

more than 300,000 firms in EU-15 countries over the period 1997-2013,

focusing in particular on the productivity of target firms as well as their

ability to establish successful market linkages. Results suggest that

selection on target firms’ profitability systematically drives MNE

strategies of cross-border takeovers: that is, domestic firms that

experience an increase in their business have a higher probability of

being acquired in any given year. By contrast, firm efficiency, in terms of

labour productivity, does not relate to international acquisition decisions,

but the effect of firm profitability tends to be concentrated in the group of

more efficient firms. These findings are confirmed also by employing

different measures of firm performance. Baseline results still hold across

a large number of checks and extensions, indicating that within-firm

differences in profitability are relevant drivers of cross-border

acquisitions.

Part III – The impact of FDI on recipient economies

Finally, building on previous chapters on the determinants of location

choices and selection patterns in cross-border takeovers, the third part of

the thesis focuses on the impact of FDI on recipient areas in terms of

24

their innovation potential. This analysis is developed as a quantitative

study having the objective of isolating the causal effect of MNE

operations on the innovative performance of host regions. In this respect,

this fifth chapter studies the extent to which knowledge externalities

arising from FDI foster local innovative performance. The quantitative

analysis is developed by employing manufacturing data on Italian

provinces over the period 2001-2006 with the specific objective of

investigating the causal impact of inward FDI on the local generation of

innovation. Adopting a Knowledge Production Function approach (KPF),

the chapter suggests that in the case of Italy the presence of foreign

investment is beneficial for the innovative performance of the recipient

local economies. These results are robust to a number of checks, thus

contributing with new evidence to the literature on the impact of FDI on

destination countries. In terms of policy consideration, this implies that a

structured policy for the attraction of external capital might channel

additional sources of knowledge to complement local capabilities.

IV. Concluding remarks

This thesis focuses on the study of MNE activities in the global

economy, providing a comprehensive and novel examination of specific

aspects of corporate operations of crucial relevance for academic and

policy purposes. In this respect, the thesis is comprehensive since it

covers both determinants and impacts of MNE activities and FDI,

considering not only the viewpoint of MNEs but also that of recipient

economies and domestic firms. The thesis also provides an original

contribution since it identifies new areas of enquiry within the vast and

well-established literature on MNEs, by asking novel research questions

and/or by combining original data sources, methodologies and

conceptual perspectives to address existing questions on which empirical

evidence remains mixed or inconclusive.

25

The three parts of the thesis are complementary in addressing various

and interconnected aspects of MNE strategies and behaviour, thus

developing and following an imaginary fil rouge that starts from the

analysis of the location decisions of MNEs undertaking greenfield FDI,

crossing the patterns of selection in the decisions of MNEs engaging in

cross-border acquisitions, and ending with the examination of the impact

of FDI on host economies’ innovative capacity at a detailed geographical

level. In general, what emerges from the various chapters is that the role

played by MNEs in the global economy is increasingly relevant and that

these actors are able to shape the patterns of international investment

and, ultimately, the trajectories of economic development at both

national and subnational level. The continuous re-organisation of

international production in response to MNE strategies and behaviour,

therefore, deserves further analysis as far as most of the aspects

addressed in this thesis are concerned, including MNE heterogeneous

preferences with respect to location-specific attributes such as economic

institutions, MNE selection strategies underpinning cross-border

takeovers, and the long-standing but still inconclusive issue of FDI-

induced localised knowledge spillovers. In this sense, this thesis

contributes to pave the way for further research on aspects of MNEs and

FDI that are in part overlooked by existing studies or subject to

conceptual and empirical controversy.

26

Figure 1: Growth of FDI, trade and GDP in the world, 1970-2010

100

200

300

400

500

600

700

800

World exports

FDI

World nominal GDP

Valu

e in

dex (1970=100)

Year

Source: own elaboration on UNCTADSTAT data.

27

Part I: MNE Location Strategies

28

Chapter 1 - The geography of foreign investments in the EU Neighbourhood

1.1 Introduction

Over the past decades the world economy has been characterised by

an increasing process of internationalisation of economic activities with

the involvement of a growing number of countries. According to

UNCTAD, the world stock of Foreign Direct Investment (FDI) in 2010 has

reached $20 trillion dollars, while the figure for the first half of the 1980s

was below one trillion.2 The dramatic expansion of international

investment represents one of the main features of the process of

globalisation, in which developing and transition economies have been

progressively more involved (e.g. Moran, 1999; Asiedu, 2002; Iammarino

and McCann, 2013).

This paper aims to explore the geographical patterns of FDI in a set of

developing and transition economies linked to the 'core' of the European

Union (EU-15) by different degrees of functional, economic and political

integration, and that will be broadly referred to as the ‘EU

neighbourhood’. Such an area embraces the EU New Member States

(NMs) that joined the EU in 2004 and 2007 (strongest degree of

integration with the 'core' of the EU-15), Accession and Candidate

Countries (ACC), European Neighbourhood Policy (ENP) countries, and

Russia (the latter with the weakest degree of integration with the EU-15,

stronger autonomy, but crucially important 'gravitation point' for

investments in the area).3 This group of countries represents a very

2 http://unctadstat.unctad.org.

3 NMs: Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, Slovakia

and Slovenia; ACC: Albania, Bosnia and Herzegovina, Croatia (which joined the EU in 2013), Macedonia,

Montenegro, Serbia and Turkey; ENP Southern: Algeria, Egypt, Israel, Jordan, Libya, Lebanon, Morocco,

Syria, Tunisia; ENP Eastern: Armenia, Azerbaijan, Belarus, Georgia, Moldova, Ukraine.

29

relevant case in terms of patterns of FDI and strategies of multinational

enterprises (MNEs) for its geographical proximity as well as its political

and economic links to the EU-15 economic core. In this respect, the

paper offers some new insights on the dynamics of global investment in

the EU neighbourhood. While this region is relatively under-explored in

the existing literature on FDI, its importance from a policy perspective is

rapidly increasing. Policy-makers at the EU and national level are

especially interested in gaining a better understanding of FDI patterns

(and their drivers): the European Neighbourhood Policy and the

intensification of economic and institutional relationships with other

important actors in the area (such as the Russian Federation and

Turkey, among others) have made apparent the huge potential of the

entire region in terms of future economic development and integration

through global value chains. Furthermore, the attractiveness of these

economies for international investment is of special interest because of

their relatively recent access to global markets that has often been

coupled with (or mediated by) a close relationship with the European

Union, making them unique case studies for the analysis of the

interaction between globalisation and regionalisation processes. As a

consequence, from the standpoint of academic research, the investigation

of MNE behaviour in terms of investment strategies in the EU

neighbourhood has a particular relevance for a better understanding of

the economic, social and geographical processes that connect global and

local actors.

This paper is based on data on individual greenfield investments in

the EU neighbourhood over the 2003-2008 period and investigates three

main aspects of the interaction between recipient countries and global

capital flows. First, the analysis aims to single out which national

characteristics are relevant for attracting global FDI into the EU

neighbourhood. Second, the paper examines the role of different FDI

30

determinants across sectors and business activities in order to shed new

light on the heterogeneous effect of different characteristics of the

recipient economies on investments of different nature. Third, the

analysis investigates whether FDI originating from different areas of the

world responds differently to national features and concentration

patterns.

The next section provides a brief overview of the empirical research

that has explored FDI determinants in the EU neighbourhood, while

Section 3 offers a detailed picture of FDI patterns in this area. Section 4

introduces the drivers of FDI considered in the econometric section and

explains the methodology. The main findings are presented and

discussed in Section 5, whilst Section 6 concludes.

1.2 Literature background: the drivers of FDI

into the EU neighbourhood

In recent years, the intensity of the political and economic relations

between the EU-15 and its neighbouring countries has increased

substantially. However, the EU relations with its neighbours have been

far from homogeneous, considering the remarkable differences among

these countries. Some ex-socialist Central and Eastern European

countries (CEECs) succeeded in joining the Union in the enlargement

rounds of 2004 and 2007, while others are still candidate to accession.

In addition, a heterogeneous group of countries geographically bordering

the EU has become part of the so-called European Neighbourhood Policy,

a unified framework aiming at generating peaceful and collaborative

relationships between the EU and its border countries (Commission of

the European Communities, 2004).

31

Notwithstanding the variety of economies involved – to different

degrees – in this process, the attention of most existing studies on FDI

and their determinants in the area has been focused on CEECs (i.e. the

countries that gained full EU membership in the 2000s and that are here

called New Member states – NMs).4 Most existing studies looked at FDI

flows in the NMs in order to understand whether (and to what extent)

increasing economic integration can influence FDI drivers. The reason for

the special attention devoted to this sub-group of countries by the

existing academic literature is threefold. First, the EU enlargement has

provided scholars with unprecedented settings for the study of FDI

patterns. Second, these analyses responded to the widespread concerns

for the growing de-localisation (and potential job loss) away from the 'old'

EU members in favour of CEECs (e.g. Boeri and Brücker, 2001). The

third reason is related to data availability: not only NMs have received a

much larger share of FDI than all other countries in the EU

neighbourhood, but empirical analyses have also been fuelled by more

accessible and comparatively more reliable data.

What emerges from the literature on the determinants of FDI in NMs

is that internal demand, market potential and labour costs are

fundamental aspects that foreign firms consider in their investment

decisions (Resmini, 2000; Carstensen and Toubal, 2004; Janicki et al.,

2004; Bellak et al., 2008). Other relevant elements for FDI attraction

include proximity to the EU (Bevan and Estrin, 2004), deepening

economic integration (Brenton et al., 1999), good institutions (Bevan et

al., 2004) and tax incentives (Bellak and Leibrecht, 2009). Interestingly

for the aims of the present paper, Resmini (2000) develops an empirical

model taking into account sectoral differences in attracting FDI in NMs:

her findings suggest that the responsiveness of FDI to national

4 As Croatia joined the EU on the 1

st of July 2013, in this paper it is considered Accession country and

included in the ACC group.

32

characteristics differs substantially across industries. This insight is

corroborated by the results of Pusterla and Resmini (2007), showing that

sector-specific drivers influence the investment decisions of foreign

companies in NMs. The present paper offers a similar perspective for

countries of the EU neighbourhood, further extending the analysis to

business functions, following Crescenzi et al. (2014).

In sharp contrast with the abundance of studies on NMs, FDI

patterns in the EU neighbourhood are much less explored in the

literature. The limited number of studies on the area converges in

suggesting that 'traditional' FDI determinants matter the most in this

context. For instance, studies on the subnational determinants of FDI in

Turkey suggest that local demand and agglomeration forces are very

relevant drivers of FDI (Deichman et al., 2003). FDI in the Balkan region

tends to be encouraged by low labour cost (Louri et al., 2000) and

political and economic reforms (Sergi, 2004). Some contributions have

investigated the determinants of FDI in the Middle East and Northern

Africa (MENA) countries, showing that growing markets, human capital

and low risk environments exert a strong attractive influence on global

investment (Moosa, 2009). The role of market size, trade opportunities

and institutional variables, along with the availability of natural

resources, is confirmed by other studies on FDI in MENA countries

(Hisarciklilar et al., 2006; Mohamed and Sidiropoulos, 2010). Recent

work by Zvirgzde et al. (2013) on Ukrainian survey data argues that FDI

in the capital region are mostly market-seeking, and also motivated by

institutional factors, while FDI in western areas are attracted by the

proximity to the EU. A strong market-oriented rationale for FDI is also

found by studies on Russia (Fabry and Zeghni, 2002; Ledayeva, 2009); in

addition, in the latter case FDI is motivated by both resource-seeking

strategies and availability of physical infrastructure such as sea ports

(Ledayeva, 2009).

33

Overall, although the literature on FDI determinants has devoted

limited attention to the EU neighbourhood, at least in comparison to

other emerging areas such as China, India or Latin America, existing

contributions point out that most FDI in the region follows market

and/or efficiency-seeking rationales.

1.3 Stylised facts on FDI in the EU

neighbourhood

In order to broaden the perspective of the existing literature and cover

both the EU NMs and the broadly defined neighbourhood of the Union

this paper makes use of homogenous and comparable data on individual

investment projects undertaken by MNEs in 34 countries in the period

2004-2008.5 The source of data is FDi Markets-Financial Times Business,

which represents an increasingly exploited tool of analysis in the

literature on FDI determinants and location choices (e.g. Crescenzi et al.,

2014).6 Greenfield investments from the entire world into the EU NMs

and neighbourhood are used to investigate country-level drivers of FDI

decisions. In what follows we present some descriptive evidence in order

to contextualize the subsequent empirical analysis.

[Table 1.1 here]

5 Although FDi Markets provides data since 2003, in the present work we consider only the period 2004-

2008. This is due to the econometric exercise requiring lagged independent variables for which data are not

available prior 2003 (see Section 4 below). 6 FDI is identified by Financial Times’ analysts through a wide variety of sources, including nearly 9,000

media sources, project data provided from over 1,000 industry organisations and investment agencies, and

data purchased from market research and publication companies. Furthermore, each project is cross-

referenced across multiple sources and more than 90% of investment projects are validated with company

sources. The dataset is by construction a sample of global FDI, and it is therefore likely to be skewed

towards the larger firms and projects. However, Crescenzi et al. (2014) show that investment decisions

captured by this database are highly correlated with other macro-level data on FDI from UNCTAD and the

World Bank.

34

As is mentioned above, the EU neighbourhood, as considered here, is

a highly heterogeneous region. NMs have joined the EU in two

subsequent enlargement rounds in 2004 and 2007, ACC are EU

candidate or potential candidate, while a large group is involved in the

ENP, with the exception of Russia. These different degrees of integration

with the EU signal the large variation in economic and political features

across the region, as well as in the extent of attractiveness towards global

capital flows.

Table 1.1 reports new foreign investments undertaken in the EU

neighbourhood over the period 2004-2008 by global MNEs. Over half of

total FDI flows in the area are directed to NMs (52.18%), while ACC, ENP

Southern and ENP Eastern economies all exhibit lower and similar

shares: 10.03%, 11.92% and 8.0%, respectively. A relevant share is,

instead, targeting Russia, which receives 18.11% of total global FDI

directed in the area. Considering individual countries rather than

groups, Russia is the most attractive destination for FDI, followed at

large distance by Romania (11.91%), Poland (9.26%) and Hungary

(7.16%). In the ACC group, Turkey and Serbia are the most preferred

destinations, with 3.87% and 2.68% respectively.

In the ENP Southern region, Morocco and Egypt play a leading role

with 2.39% and 2.25% of total FDI, whilst in the ENP Eastern region

Ukraine attracts the great majority of investments with 4.67% of the

total. Figure 1 provides a graphical representation of global FDI

distribution in the EU neighbourhood over the period 2004-2008.

[Figure 1.1 here]

There are different motives behind investment decisions and they are

intimately connected to the functions and sectors in which MNEs operate

their foreign activities. Although the original dataset reports several

typologies of business functions and a large number of industrial sectors,

35

due to the low number of observations in some countries for certain

activities and industries, data are aggregated into three groups of

business functions and two broad economic sectors. With respect to the

former, Table 1.2 presents figures on investment in the following broad

functional categories: (i) Headquarter and Innovation activities (HQ &

Inno); (ii) Sales, Marketing, Logistic and Distribution (SMLD) and (iii)

Production. Table 1.3 instead provides an outlook on the macrosectoral

aggregations: (i) Manufacturing and (ii) Services.

Table 1.2 shows that NMs attract the large majority of FDI in all

business functions. However, Russia remains the single most important

country in terms of attractiveness across all functions. Surprisingly, ENP

Southern countries receive a relatively large share of FDI in

Headquarters and Innovative activities (16.7%), due in particular to the

large role played by Israel (3.8%). Among NMs, Romania attracts the

largest share of FDI in all business functions, while Turkey and Serbia

lead the ACC group. As far as ENP Eastern is concerned, Ukraine

unsurprisingly plays the most relevant role. What emerges from these

figures is that global FDI tends to be concentrated in a few locations

across the EU neighbourhood, and that variations in foreign investors’

preferences exist according to different business functions. For instance,

Poland is one of the main destinations of global FDI in the area, but only

5.9% is in Headquarters and Innovation, while the share almost doubles

when looking at FDI in Production activities.

[Table 1.2 here]

Table 1.3 reports the distribution of FDI towards the EU

neighbourhood for the two industrial macro-aggregates, which also show

remarkable differences. FDI in manufacturing concentrates in NMs

(56.3%), whilst the attractiveness of ENP Eastern, ENP Southern and

ACC groups in this respect is relatively weak (5.8%, 8.7% and 9.5%,

36

respectively); the Russian Federation alone attracts 19.7% of

manufacturing FDI. As far as service activities are concerned, the shares

of ENP Southern and ENP Eastern are higher (14.8% and 9.9%

respectively) while NMs still attract about half the volume of service FDI

(47.9%).

[Table 1.3 here]

1.4 FDI in the EU neighbourhood: methodology

In order to investigate the role (and relative importance) of national

characteristics for the attraction of FDI in the EU neighbourhood, this

paper relies upon regression techniques. In particular, following the

literature on the quantitative analysis of MNE location, the empirical

analysis relies on a count data model where national characteristics

explain the number of FDI projects received by each country in each

year.7 With a count response variable, it is customary to employ a

Poisson regression technique. However, we detect over-dispersion in our

count variable, which makes this methodology less appropriate: we

therefore apply a negative binomial model, which allows us to adjust

estimates for over-dispersed data8 9. The time span covers the period

2004-2008 and includes a total of 11,262 greenfield FDI. In line with the

relevant literature, independent variables enter the analysis with a one-

year lag, as specified below. Thus, data for 2003 are employed to

construct lagged explanatory variables.

7 Alternatively, a conditional logit model can be adopted, as common in similar studies. Nevertheless, the

equivalence of the coefficients provided by these classes of models is well established in the literature

(Guimarães et al., 2003). 8 An additional problem with count data models can derive from the large number of zeros in the data.

However, this is not a relevant issue in our dataset. 9 We also run a Poisson regression (not reported here) which confirmed the main results of the Negative

Binomial.

37

The following empirical model is estimated:

𝐹𝐷𝐼𝑖𝑡 = 𝑓(𝑑𝑒𝑚𝑎𝑛𝑑𝑖𝑡−1, 𝑖𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛𝑠𝑖𝑡−1, 𝑙𝑎𝑏𝑜𝑢𝑟𝑖𝑡−1, 𝑐𝑜𝑙𝑜𝑐𝑎𝑡𝑖𝑜𝑛𝑖 , 𝑃𝑖)

Where:

FDIit is the count of foreign investment in destination country i in year

t.

Demandit-1 represents internal market size and external market

potential (MP) of country i in year t-1; both variables enter the model in

log form. The size of the market in the host economies is viewed as a

major driver of FDI (e.g. Wheeler and Mody, 1992; Billington, 1999). The

larger the national market in the recipient country, the larger the local

demand for goods and services and, consequently, market opportunities

for the investor. National GDP at constant prices (US dollars 2005) is

included as a proxy, with one-year lag, and comes from the World

Development Indicators (WDI) of the World Bank.

FDI might also be aimed at exploiting external market potential (e.g.

Head and Mayer, 2004; Carstensen and Toubal, 2004): in other words,

some countries can play the role of platforms for exports towards other

proximate locations. In order to control for countries’ external market

potential we follow the literature (Harris, 1954) and compute the

following indicator:

𝑀𝑃𝑖𝑡−1 = ∑ (𝐺𝐷𝑃𝑐

𝑑𝑖𝑐⁄ )

𝑐≠𝑖

where market potential (MP) of location i is the distance-weighted

internal demand of neighbouring countries c. This indicator is included

in the analysis with a one-year lag.

38

Institutionsit-1 stands for ‘Control of corruption’ in country i in year t-1.

This part of the model tests whether FDI is sensitive to national

institutional environments, which are highly heterogeneous in the EU

neighbourhood. Institutions are proxied with a measure that captures a

very relevant aspect of the national environment when considering the

strategies of foreign investors, namely ‘Control of corruption’ as provided

by the World Bank in its World Governance Indicators (WGI). As for

previous variables, institutions enter the analysis with a one-year lag. As

is suggested by the existing literature, we expect that good institutional

quality plays a positive role in attracting foreign capital since it increases

certainty in market transactions and stability (e.g. Altomonte, 2000; Wei,

2000; Bénassy Quéré et al., 2007).

Labourit-1 includes proxies for the education level and average wage in

country i in year t-1. This section of the model looks at the

characteristics of the workforce and labour market. First, a measure of

the average education level in the host economy is included, that is the

ratio between secondary school age population and total population

provided by UNESCO. This is the only available measure of education for

the countries of interest. In line with studies highlighting the beneficial

effects of human capital on FDI attraction, we expect that this indicator

is positively linked to inward FDI (Noorbakhsh et al., 2001). Second, we

include per capita GDP as a proxy for average wage employing data on

GDP and population from WDI (Alsan et al., 2006). Although this is an

indirect measure for salaries, wages for most countries under

observation are not available. We expect that higher values of this

indicator discourage foreign investors, since saving on input costs

represents a strong rationale for FDI in emerging and developing

economies (Resmini, 2000).

39

Colocationit includes several stock variables for FDI in country i

calculated as a cumulative count according to country of origin, sector

and business function, all expressed in log. These variables capture the

extent to which foreign investments co-locate in the same country; that

is, using data at the investment level, we generate the stock of all FDI

with similar characteristics to those of each specific investment (e.g.

Defever, 2006). Then, when constructing our dataset at the country level,

we consider the cumulative average stock of FDI in a specific country in

a specific year. The FDi Markets database allows constructing stock

measures of FDI according to (i) nationality of the investor, (ii) sector and

(iii) business function. We are thus able to investigate the importance of

similar FDI in determining new flows of investment, exploring FDI path-

dependency along these three different dimensions. Similarly, two

additional stock variables are built by crossing both sectors and business

functions with information on origin countries, allowing to test whether

FDI in one sector or business activity originated from a certain country

attracts more FDI with similar features.

Finally, Pi is a set of country dummies included in order to account

for any factor not explicitly controlled for in the model that might have an

effect on countries’ attractiveness towards global FDI. These include any

time-invariant country-level driver of FDI such as geographical and

cultural characteristics. The full list of variables is reported in Appendix

A.

1.5 Results

The first objective of our empirical exercise is to analyse the relevance

of different FDI determinants in the EU neighbourhood. Therefore, we

estimate a negative binomial model by including all FDI directed towards

the 34 countries in the area of interest over the period 2004-2008.

40

[Table 1.4 here]

The results of this first estimation are reported in Table 1.4. The

coefficients are mostly in line with expectations, and consistent across

different model specifications. Traditional drivers of FDI, such as size of

the internal market and external market potential, are strongly and

positively correlated with the decision to undertake new investments.

This confirms that global FDI flows towards the EU NMs and

neighbourhood have a prominent market-seeking rationale. In other

words, MNE strategies in the area are strongly based upon market access

considerations in terms of both the exploitation of domestic demand in

the recipient economies and the opportunity to constitute platforms for

exports towards third countries (see Neary, 2007). As far as the national

institutional environment is concerned, ‘Control of corruption’ exhibits a

positive and weakly significant relationship with FDI in only two

specifications out of five: overall, according to this first set of results,

global investors do not appear overly concerned about choosing locations

where the institutional setting confers stability to their operations and

transactions.

With respect to workforce characteristics, the model does not detect

any relevant relationship between FDI and education level, indicating

that, in general, MNEs do not invest in the EU neighbourhood in order to

take advantage of local competences. Conversely, our proxy for wage

levels reveals that investors look for cheap labour in the region. The

robustness of the coefficient on this feature across all specifications

suggests an efficiency-seeking rationale for foreign companies investing

in the area. This indicates that the conclusions reached by previous

studies arguing that cost-saving on labour is among the main drivers for

FDI in CEECs (Resmini, 2000) may be extended to the broader EU

neighbourhood.

41

As far as FDI path-dependency is concerned, we enter the different

colocation variables separately given the high level of correlation among

them. The first three columns test the relevance of colocation patterns

associated with common nationality of the investor, sector and business

function respectively. Columns 4 and 5, instead, test the effect of

colocation of FDI in the same sector and business by nationality. Results

in Table 1.4 suggest that FDI tends to follow previous investment flows

with similar features, with the only exception of functional colocation.

Moreover, regressions in columns 4 and 5 indicate that FDI from the

same country of origin tends to select the same location according to

their sector and business activity performed abroad.

Foreign investment might be motivated by different determinants

depending on the specific function operated abroad or the particular

sector in which the FDI is undertaken. Therefore, we run separate

regressions for the three types of business functions (Table 1.5) and the

two macro-aggregates of economic activity (Table 1.6).

[Table 1.5 here]

As is shown in Table 1.5, when considering the number of FDI in

specific business functions as response variable, FDI patterns are

significantly associated with a smaller number of determinants, which

are particularly important for a specific function. Therefore, in the case of

‘HQ & Inno’, the education level of countries appears to be the main

relevant driver of FDI. This is not surprising considering that activities in

‘HQ and Inno’ are likely to be related to higher skill-intensity. Conversely,

in the case of ‘SMLD’ results suggest that a lower level of education is

attractive of FDI, plausibly signalling that these activities require less

skilled workers. As far as Production activities are concerned, a

42

favourable institutional environment plays a major role in driving FDI

patterns in the EU neighbourhood. With respect to colocation variables,

path-dependency in FDI inflows emerges clearly in the case Production.

This is not surprising considering that production activities are likely to

be associated with the occurrence of agglomeration economies and

localised backward and forward linkages. However, in the case of ‘HQ &

Inno’ the coefficients turn out to be negative and significant: this might

be due to the fact that, while corporate headquarters tend to concentrate

in large urban agglomerations (particularly capital cities) mainly for

political networking and lobbying reasons, this is not normally the case

for innovation activities (Iammarino and McCann, 2015). Previous

research has shown that MNE technological and innovation operations

are unlikely to be located in the vicinity of those of competitive rivals

(see, among others, Cantwell and Santangelo, 1999; Alcácer, 2006;

Verbeke et al., 2009) and tend rather to follow the location of production

operations (Defever, 2006) or to reflect a value chain logic (Crescenzi et

al. 2014)

[Table 1.6 here]

Table 1.6 presents results of negative binomial estimates by

macrosector. Interestingly, and not entirely unexpectedly, the signs of

the significant coefficients are opposite in manufacturing and services, a

plausible outcome in the set of countries that constitute the EU

neighbourhood. As far as manufacturing industries are concerned, the

strong and negative significance of the education level signals that

foreign MNEs tend to look for low-skilled workforce, reasonably because

the kind of manufacturing activities localised in the EU neighbourhood

by MNEs is mostly concentrated in the more basic segments of the value

chain. Differently, service activities are associated with a more educated

workforce in relation to the nature itself of the service sector, which

43

requires relative higher standards of skills than basic manufacturing.

Table 1.6 also suggests that the institutional setting of the host countries

matters for FDI decisions, again with different signs in the two

aggregates considered. In particular, manufacturing activities are

associated with less favourable institutional conditions: this, particularly

in the case of emerging and developing economies such as those in the

EU neighbourhood, might be explained by cross industry heterogeneity

in MNEs’ preferences over institutional attributes. In other words, it has

been argued that some MNEs tend to prefer locations with weaker

economic institutions because they aim at bypassing transparent market

mechanisms in their operations abroad (e.g. Helmann, 1998; Helmann et

al., 2000; Glaeser and Shleifer, 2002; Sonin, 2003). Indeed, weaker

institutions might facilitate rent-seeking or moral hazard behaviour, or

simply allow capturing a share of host countries’ public resources,

through lobbying, subsidies or less legalized channels – such as, in the

case here, corruption. Such MNE behaviours has proved to differ across

sectors and functions: previous research has shown that MNEs in high

or medium technology manufacturing choose to locate in places where

the institutional environment is more adequately protected, while MNEs

operating in low-technology and less sophisticated sectors may consider

strong regulation in business as an obstacle.10 Hence, mechanisms of

institutional subversion (Helmann, 1998) might be easily reflected in our

results for manufacturing considering the highly heterogeneous group of

countries analysed, that include both transition and developing

countries, often characterised by notable institutional flaws. On the

contrary, the institutional environment takes the expected positive sign

when the analysis shifts to FDI in services, which include operations

aiming to provide financial and business services, soft infrastructure and

more knowledge-intensive content activities – as also the attractiveness

10

To be noted that our manufacturing aggregate includes also extraction and processing of coal, oil and

natural gas, which may prove particularly reactive to less regulated institutional settings.

44

of stronger human capital in the sector seems to point out – and that

tend to take into consideration business regulation, transparency and

enforcement of contracts as pre-requisites for their location.

1.6 Conclusions

This paper aimed at providing a first investigation of the drivers of

global FDI in the broadly defined EU neighbourhood. The area

constitutes an interesting case in terms of attractiveness towards global

MNE investments, both for its geographical closeness and its political

and economic linkages with the ‘core’ of the European Union. The

different degrees of integration with the EU, and the relatively recent

access of most neighbourhood countries to global markets, reflect their

large heterogeneity in terms of economic, social and political

characteristics, which also entails large variation in their attractiveness

towards foreign capital.

By employing data on greenfield investment projects occurred in the

EU NMs and neighbourhood in the period 2003 to 2008, we explored the

drivers of FDI by sector and business function. What emerges from the

general empirical analysis is a clear market-seeking and efficiency-

oriented rationale behind FDI in the EU neighbourhood. Interestingly,

strong co-location patterns of FDI appear along different axes – national

origin of the investor, industrial sector, and business function –

supporting the existence of path-dependency, cumulative causation

mechanisms and possible virtuous (or vicious) cycles in the impact of

globalisation on the EU neighbourhood.

The findings of this paper are largely in line with previous empirical

evidence highlighting the significance of global capital flows towards EU

NMs as compared to other areas in the EU neighbourhood. In fact, EU

NMs are characterised by large and growing internal demand, a

comparatively stable institutional environment, and relatively low labour

45

costs. Most importantly from a political point of view, they benefit from

the EU membership. However, Russia is the single country that receives

most foreign investment in manufacturing and services, plausibly due to

the relevance of its huge internal demand for MNEs’ strategies.

In interpreting our empirical results and the descriptive evidence

presented, we notice that the rest of the EU neighbourhood tends to

remain peripheral in the strategies of MNEs, with few exceptions

represented by countries such as Turkey and Ukraine, and to a lesser

extent, Egypt and Morocco. These economies are far less integrated both

politically and economically with the ‘core’ of the EU, but they are central

economic actors in their regions and it is likely that MNEs oriented

towards the exploitation of new markets and low-cost labour force will

look at them with growing interest.

The present study provides an initial investigation of the patterns of

FDI in the EU neighbourhood which can be informative for policy makers

at the EU, national and regional levels in both areas. The growing

importance of the ENP and the intensification of the economic and

institutional relationships between the EU and other important actors in

the area, such as the Russian Federation, Turkey, the Balkans and the

economies in North Africa, should be accompanied by a better

understanding of the economic processes at work. In this respect, the

evidence about the role of internal markets of destination and the

educational levels of the workforce in attracting FDI can be framed within

national and EU-wide regional and industrial policies to encourage, on

the one hand, the internationalisation of European firms – particularly

those in the current EU periphery – towards their neighbours and, on the

other, the upgrading of skills and capabilities in the recipient economies.

Policies supporting human capital and skill formation and training – at

different educational levels – are indeed crucial not only to spur

technological and innovation progress in the neighbourhood, but also to

support shifts to higher value-added activities and skill renewal

46

potentially offered by offshoring to the EU peripheral regions

geographically closer to the ENP area. Furthermore, improving

institutional quality in the neighbourhood is imperative in order to

reduce rent-seeking and inefficiencies that are detrimental to the host

economies, and tend to increase internal inequality through the

reinforcement of the dominant elites: enhancing the quality of

institutions may also attract more sophisticated activities and reduce the

current emphasis on purely market-seeking investments. Further

research-based evidence is certainly needed to inform policy intervention

on which specific tools are best suited to leverage global flows to upgrade

local tangible and intangible assets and reinforce regional growth on both

sides of the EU border.

47

Table 1.1: FDI into the EU neighbourhood

Country Investment projects %

New Member States

Bulgaria 735 6.53

Czech Republic 651 5.78

Estonia 207 1.84

Hungary 806 7.16

Latvia 293 2.60

Lithuania 236 2.10

Malta 8 0.07

Poland 1,043 9.26

Romania 1,341 11.91

Slovakia 446 3.96

Slovenia 109 0.97

Subtotal 5,875 52.18

Accession and Candidate countries

Albania 49 0.44

Bosnia and H. 96 0.85

Croatia 183 1.62

Macedonia 45 0.40

Montenegro 19 0.17

Serbia 302 2.68

Turkey 436 3.87

Subtotal 1,130 10.03

ENP Southern countries

Algeria 208 1.85

Egypt 253 2.25

Israel 120 1.07

Jordan 111 0.99

Lebanon 66 0.59

Libya 88 0.78

Morocco 269 2.39

Syria 88 0.78

Tunisia 137 1.22

Subtotal 1,340 11.92

ENP Eastern countries

Armenia 47 0.42

Azerbaijan 113 1.00

Belarus 80 0.71

Georgia 69 0.61

Moldova 43 0.38

Ukraine 526 4.67

Subtotal 878 8.00

Russia 2,039 18.11

Total 11,262 100

Source: Authors' elaborations on FDi-Markets data

48

Table 1.2: FDI into the EU neighbourhood by business function

Country HQ & Innovation SMLD Production

Investment % Investment % Investment %

New Member States

Bulgaria 82 4.5 328 6.9 325 6.9

Czech Republic 101 5.6 271 5.7 279 5.9

Estonia 34 1.9 103 2.2 70 1.5

Hungary 118 6.6 349 7.3 339 7.2

Latvia 25 1.4 191 4.0 77 1.6

Lithuania 28 1.6 153 3.2 55 1.2

Malta 1 0.06 3 0.06 4 0.08

Poland 107 5.9 394 8.3 542 11.5

Romania 223 12.4 568 12.0 550 11.7

Slovakia 48 2.7 159 3.4 239 5.1

Slovenia 14 0.8 65 1.4 30 0.6

Subtotal 781 43.1 2,584 59.4 2,510 53.3

Accession and Candidate countries

Albania 9 0.5 19 0.4 21 0.5

Bosnia and H. 13 0.7 32 0.7 51 1.1

Croatia 16 0.9 94 2.0 73 1.6

Macedonia 3 0.2 9 0.2 33 0.7

Montenegro 1 0.06 8 0.2 10 0.2

Serbia 52 2.9 119 2.5 131 2.8

Turkey 91 5.1 171 3.6 174 3.7

Subtotal 185 10.2 452 10.4 493 10.5

ENP Southern countries

Algeria 50 2.8 77 1.6 81 1.7

Egypt 43 2.4 91 1.9 119 2.5

Israel 69 3.8 30 0.6 21 0.5

Jordan 23 1.3 44 0.9 44 0.9

Lebanon 15 1.3 33 0.7 18 0.4

Libya 18 1.0 18 0.4 52 1.1

Morocco 33 1.83 104 2.2 132 2.8

Syria 20 1.1 18 0.4 50 1.1

Tunisia 32 1.8 33 0.7 72 1.5

Subtotal 303 16.7 448 10.3 589 12.5

ENP Eastern countries

Armenia 19 1.1 14 0.4 14 0.3

Azerbaijan 32 1.8 50 1.1 31 0.7

Belarus 19 1.1 45 1.0 16 0.3

Georgia 17 0.9 32 0.7 20 0.4

Ukraine 132 6.5 237 5.0 168 3.6

Moldova 4 0.2 14 0.3 14 0.3

Subtotal 223 12.3 392 9.0 263 5.6

Russia 319 17.6 866 19.9 854 18.1

Total 1,811 100 4,350 100 4,709 100

Source: Authors' elaborations on FDi-Markets data

49

Table 1.3: FDI into the EU neighbourhood by macro-sector

Country Manufacturing Services

Investment % Investment %

New Member States

Bulgaria 358 6.0 323 6.8

Czech Republic 401 6.7 226 4.8

Estonia 112 1.9 90 1.9

Hungary 476 7.9 292 6.2

Latvia 174 2.9 117 2.5

Lithuania 125 2.1 100 2.1

Malta 6 0.1 2 0.04

Poland 605 10.1 413 8.7

Romania 748 12.5 552 11.6

Slovakia 310 5.2 125 2.6

Slovenia 59 1.0 43 0.9

Subtotal 3,374 56.3 2,283 47.9

Accession and Candidate countries

Albania 18 0.3 23 0.5

Bosnia and H. 48 0.8 48 0.8

Croatia 100 1.7 100 1.7

Macedonia 16 0.3 19 0.3

Montenegro 3 0.05 3 0.05

Serbia 171 2.9 122 2.6

Turkey 214 3.6 200 4.2

Subtotal 570 9.5 515 10.8

ENP Southern countries

Algeria 89 1.5 102 2.2

Egypt 102 1.7 127 2.7

Israel 49 0.8 65 1.4

Jordan 44 0.7 65 1.4

Lebanon 18 0.3 47 1.0

Libya 21 0.4 39 0.8

Morocco 108 1.8 152 3.2

Syria 25 0.4 48 1.0

Tunisia 68 1.1 61 1.3

Subtotal 524 8.7 706 14.8

ENP Eastern countries

Armenia 14 0.2 26 0.6

Azerbaijan 35 0.6 64 1.4

Belarus 31 0.5 46 1.0

Georgia 17 0.3 39 0.8

Moldova 19 0.3 20 0.4

Ukraine 229 3.8 276 5.8

Subtotal 345 5.8 471 9.9

Russia 1,180 19.7 792 16.7

Total 5,993 100 4,767 100

Source: Authors' elaborations on FDi-Markets data

50

Table 1.4: FDI determinants into the EU neighbourhood

(1) (2) (3) (4) (5)

Dep Var: FDI count

Market size 2.80*** 2.89*** 2.74*** 3.21*** 3.11***

(0.909) (0.936) (0.917) (0.846) (0.866)

Market potential 2.64** 2.62** 2.91*** 2.12** 2.47**

(1.103) (1.124) (1.094) (0.999) (1.027)

Control of corruption 0.47* 0.43 0.44 0.39 0.44*

(0.273) (0.274) (0.278) (0.248) (0.260)

Education level 1.28 1.33 1.28 1.11 1.27

(0.848) (0.876) (0.890) (0.757) (0.786)

Average wage -3.15*** -3.18*** -3.10*** -3.49*** -3.53***

(0.863) (0.879) (0.874) (0.803) (0.811)

National colocation 0.004**

(0.0016)

Sector colocation

0.004**

(0.00214)

Function colocation

0.001

(0.000781)

Sector colocation by nationality

0.062***

(0.0124)

Function colocation by

nationality

0.027***

(0.00660)

Observations 170 170 170 170 170

National dummies Yes Yes Yes Yes Yes

Pseudo R-squared 0.28 0.28 0.28 0.30 0.29

log likelihood -573.4 -573.8 -574.7 -564.7 -569.1

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

51

Table 1.5: FDI determinants in the EU neighbourhood by business function

(1) (2) (3) (4) (5) (6) (7) (8) (9)

Dep Var: FDI count HQ & Inno SMLD Production

Market size 9.11 8.15 8.90 -1.11 -1.16 -1.29 -0.96 -0.087 -0.37

(6.577) (6.321) (6.500) (5.929) (6.122) (6.273) (3.141) (3.187) (3.156)

Market potential -1.21 -1.24 -2.84 -5.77 -5.87 -6.07 1.20 0.10 0.65

(5.315) (5.179) (5.285) (6.632) (6.911) (6.949) (3.552) (3.484) (3.451)

Control of corruption 0.56 0.69 0.44 -1.02 -0.91 -0.92 2.27** 2.10** 2.22**

(1.323) (1.334) (1.328) (0.995) (0.986) (0.987) (0.992) (1.001) (0.998)

Education level 14.24*** 15.19*** 14.25*** -3.60** -3.64** -3.74** 3.11 4.88 5.17

(4.476) (4.775) (4.580) (1.624) (1.639) (1.648) (3.588) (3.624) (3.555)

Average wage 6.36 9.57 9.39 2.71 2.56 2.77 0.43 -0.05 -0.09

(6.390) (7.011) (7.111) (3.785) (3.823) (3.903) (2.307) (2.312) (2.330)

National colocation -0.02

-0.01

0.01

(0.012)

(0.009)

(0.010)

Sector colocation

-0.04**

-0.01

0.025*

(0.02)

(0.011)

(0.014)

Function colocation

-.015***

-0.002

0.011**

(0.005)

(0.003)

(0.005)

Observations 170 170 170 170 170 170 170 170 170

National dummies Yes Yes Yes Yes Yes Yes Yes Yes Yes

Pseudo R2 0.28 0.30 0.30 0.16 0.16 0.16 0.15 0.16 0.16

log likelihood -56.40 -55.30 -55.34 -100.1 -100.2 -100.2 -95.21 -94.57 -94.38

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

52

Table 1.6: FDI determinants in the EU neighbourhood by macro-sector

(1) (2) (3) (4) (5) (6)

Dep Var: FDI count Manufacturing Services

Market size -1.96 -1.63 -1.61 4.06 4.13 4.26

(3.725) (3.737) (3.720) (3.832) (3.688) (3.683)

Market potential -2.37 -2.92 -2.91 0.43 0.45 0.11

(3.745) (3.755) (3.639) (3.154) (3.059) (3.106)

Control of corruption -3.19*** -3.16*** -3.15*** 1.55** 1.51** 1.46*

(0.923) (0.930) (0.933) (0.776) (0.750) (0.754)

Education level -5.00*** -4.75** -4.71** 4.22** 4.33** 4.28**

(1.919) (1.983) (2.000) (2.012) (2.016) (2.015)

Average wage 0.67 0.47 0.44 -1.93 -1.49 -1.15

(2.374) (2.365) (2.385) (3.157) (3.133) (3.155)

National colocation -0.003

-.0004

(0.007)

(0.010)

Sector colocation

0.001

-0.008

(0.009)

(0.012)

Function colocation

0.0004

-0.004

(0.003)

(0.004)

Observations 170 170 170 170 170 170

National dummies Yes Yes Yes Yes Yes Yes

Pseudo R2 0.16 0.16 0.16 0.12 0.12 0.12

log likelihood -104.4 -104.4 -104.4 -107.9 -107.8 -107.7

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, *p<0.1

53

Appendix A

Table A.1: List of variables

Variable Description Source

Dependent

FDIit Count of FDI in country i at time t FDi

Markets

Independent

Demand

Market Sizeit-1 GDP of country i at time t-1. WDI

Market Potentialit-1 Sum of distance-weighted GDP of all third countries c from location i at time t-1.

WDI / CEPII

Institutions

Control of Corruptionit-1 Composite indicator ranging from -2.5 to 2.5,

with higher values associated to more control of corruption in country i at time t-1.

WGI

Labour

Education Levelit Ratio between secondary school age population and total population in country i at time t-1.

UNESCO

Average Wageit Per capita GDP in country i at time t-1. WDI

Co-location

National Co-locationit Cumulative average stock of investment in country i from the same country of origin.

FDi

Markets

Sector Co-locationit Cumulative average stock of investment in country i in the same sector of activity.

FDi

Markets

Function Co-locationit

Sector Co-locationit by

nationality

Function Co-locationit by

nationality

Cumulative average stock of investment in country i in the same business function.

Cumulative average stock of investment in country i in the same sector of activity from the

same country of origin.

Cumulative average stock of investment in country i in the same business function from the

same country of origin.

FDi Markets

FDi

Markets

FDi

Markets

54

Chapter 2 – What drives European multinationals to the EU neighbouring countries? A mixed methods analysis of Italian investment strategies

2.1 Introduction

The progressive enlargement of the European Union (EU) has made

the economic and political relationships with its neighbours a highly

sensitive policy issue. With the EU Enlargement the security, political

stability and economic prosperity of larger shares of the Union are

progressively more intertwined with that of Candidate and Neighbouring

countries. Following the 2004 and 2007 eastward enlargements, the

European Neighbourhood Policy (ENP) and other regional and multi-

lateral cooperation initiatives (Eastern Partnership; the Euro-

Mediterranean Partnership; the Black Sea Synergy and the EU-Russia

strategic partnership) have been aimed at strengthening the links

between the EU and its neighbourhood in institutional, political, social

and economic terms. The sharp increase in trade flows (according to the

European Commission total trade between the EU and its ENP partners

was worth € 230 billion in 2011) and labour mobility (the EU issued 3.2

million Schengen visas to ENP partners in 2012) has been accompanied

by a generalized increase in Foreign Investments in particular towards

the ENP-South countries. Before the 2007 economic crisis, FDI in the

Mediterranean region accounted for 2.8% of the world total (2006) while

investments in Eastern countries remained largely concentrated in

Ukraine, ranging between 0.5 and 1% of the world total (DRN, 2013): the

EU accounts on average for 34% of total investments in the

Mediterranean countries (while no comparable data are available for

55

Eastern countries, but EU FDI account for around 80% of the total in

Ukraine) (DRN, 2013).

While “corruption has been identified as a major obstacle to

investment and business, both in eastern and southern ENP countries”

(European Commission, 2013: 10), very limited systematic research has

been conducted so far on the relative importance of other investment

drivers/barriers that might play an important role in this emerging

context. Corruption and poor institutional quality remain fundamental

cross-country issues for the entire region (see Chapter 3 for a more

detailed discussion of this), but market-seeking (associated with

increasing market size), resource-seeking and efficiency-seeking

(associated with cheap skilled labour) motives remain strong

countervailing pull factors that interact with geographical and

(increasing) institutional proximity, sustaining the increasing flow of EU

investments in the region.

This paper aims to shed new light on the strategic decisions of

European MNEs when balancing the repulsive and attractive forces that

shape the geography of their investments in the EU neighbouring

countries (NCs) and in the ‘new’ member states (NMs) of the EU. The

coverage of 33 destination countries among NCs and European NMs11

makes it possible to analyse the full spectrum of economic and

institutional integration with the ‘core’ of the EU-15, from the full

economic and political integration into the Union and the single market

of the NMs, to the looser association of the ENP East and South. In terms

of origin of the investments the focus of the paper will be on the case of

Italy. The focus on investments originating from one single country will

make possible to ‘net out’ any ‘home market’ bias in MNE behaviour,

11

In this paper NCs are (i) Accession and Candidate Countries (ACC): Albania, Bosnia and Herzegovina,

Croatia, Macedonia, Montenegro, Serbia and Turkey; (ii) ENP Southern countries: Algeria, Egypt, Israel,

Libya, Lebanon, Morocco, Syria and Tunisia; (iii) ENP Eastern countries: Armenia, Azerbaijan, Belarus,

Georgia, Moldova and Ukraine; and (iv) Russian Federation. EU NMs are all 2004 and 2007 European

enlargement countries except Cyprus.

56

allowing us to compare their strategies with reference to the highly

diversified context of the NCs and NMs. The case of Italy is particularly

appropriate for this purpose: Italy is a founding member of the European

Union that forms part of the ‘core’ of the Union but, at the same time,

benefits from closer geographical proximity with both NMs and NCs than

other ‘old’ EU members. In addition, Italian foreign and commercial

policies have historically devoted a special attention to the role of the

country as a ‘bridge’ between the ‘Old’ Europe and the EU neighbourhood

(Bank of Italy, 2000)

The analysis of investment strategies in both NMs and NCs needs to

take into account not only the variety of contextual conditions of the host

economies but also the diversity of the entry modes of foreign firms into

the local markets (European Commission, 2014). As a consequence, this

paper will adopt a mixed methods approach to the analysis of the

location strategies of Italian investments in the area. Drawing on

Dunning’s Ownership-Localization-Internalization (OLI) eclectic

paradigm, the paper uses regression analysis in order to assess the

different role of national drivers in affecting Italian greenfield

investments’ location behaviour. This section of the analysis is based on

detailed data at the level of individual investment project. However, in

order to capture the complex interaction between greenfield investments

and other entry modes (in particular joint ventures and acquisitions) the

quantitative analysis is complemented by two in-depth firm-level case

studies covering two of the largest Italian multinational enterprises

operating with different modalities in both the EU NMs and NCs areas.

Interviews are collected at the level of headquarters with top level

managers and executives, presenting a rich informative basis on the

strategic behaviour and organisational choices of MNEs in their cross-

border operations in NCs and NMs.

57

In terms of contribution to the existing debate, the paper rests on the

idea that MNE investments play a central role in the on-going process of

integration between the EU and its neighbouring countries. Such a

critical role has been rarely investigated with mixed methodologies,

which instead offer the opportunity to analyze more in-depth the

interaction between patterns of economic integration and business

strategies of MNEs. Therefore, the contribution of the present study is

essentially empirical. In this respect, the paper aims at providing a

structured analysis of associations between recipient countries’

attributes and corporate behavior in the quantitative part, fundamentally

assessing the role of location advantages (L) of the eclectic OLI paradigm

to motivate Italian MNEs to pursue internalization (I) strategies.

Subsequently, the qualitative section of the article zooms into the

investment behavior of two selected Italian multinationals, capturing the

full complexity that is typical of MNE organizational choices and that is

rarely incorporated in existing quantitative studies. In this respect, we

are also able to explore MNE characteristics as drivers of their location

choices, with the aim of capturing the forms of ownership advantages (O)

that lead to internalization (I). Therefore, by combining quantitative and

qualitative insights in a novel way, this article provides new empirical

evidence on the location strategies of MNEs taking into account the

interdependence between the different components of Dunning’s OLI

paradigm, that is destination country determinants and firm-level

organizational features that drive cross-border corporate strategies.

The main findings of the mixed-methods analysis for Italian MNEs in

the EU neighbourhood suggest that, while some common elements for

localisation – such as market access considerations as well as sensitivity

to cost factors – can be generalised, there is evidence of an intrinsic

heterogeneity in the strategies of MNEs along sector and functional axes,

ranging from the role of inter-governmental agreements to the

importance of institutional assimilation of the MNE in the local context.

58

This diversity across corporate strategies suggests that the development

of ‘framework conditions’ within the picture of further integration

between the EU and its neighbourhood is at least as important as the

reinforcement of more typical FDI attractors.

The paper is organised as follows. The next section briefly outlines the

characteristics of Italian foreign investment in EU NMs and NCs. Section

3 introduces the quantitative analysis of Italian MNEs location strategies:

the empirical model is presented and justified and the results of the

regression analysis are discussed. Section 4 briefly introduces the

corporate profile of the Italian MNEs covered in the study, whilst section

5 analyses the evidence from the in-depth interviews with the executives.

Section 6 concludes.

2.2 Italian Foreign Investments in EU New

Member States and Neighbouring Countries

Italy is a key player in global investments towards the EU NMs and

NCs. According to the International Monetary Fund (IMF) Italy’s global

outward investment has reached $535 billion in 2012 with $69.42 billion

(approximately 13% of the total) going to the area of interest for this

paper, suggesting that the region is extremely relevant to Italian foreign

operations. Table 2.1 shows Italian investments in the countries of the

area combining information from the Coordinated Direct Investment

Survey of the International Monetary Fund12 in the most recent available

year with data on Italian new investment projects in the period 2003-

2008 from the FDi Markets database created by Financial Times

Business13. IMF macro-economic FDI data provide us with a complete

12

http://cdis.imf.org/ 13

FDi Markets is the leading source of information on Foreign Direct Investments, providing data to the

UNCTAD report and the World Bank. For each project detailed information is available on the investor

(name and state/country of origin and sector of activity, including both manufacturing and services), on the

destination area (country, state and city), and the main business function (including manufacturing, sales

and marketing, R&D, logistic, headquarter and business services) involved in the investment abroad.

59

and updated picture of all Italian FDI flows in the area. However, IMF

data are only available after 2009 and they do not include any detail on

the nature of the investments. Conversely, FDi Markets data include

detailed micro-level information on new foreign investment project

undertaken in the region with sector and function breakdown based on

the combination of a variety of local and media sources. The two data

sources are highly correlated (65% correlation for the individual

countries’ shares of total investments; 93% correlation for the regional

sub-totals reported in Table 2.1) confirming that FDi Markets micro-data

– used here in the quantitative analysis – offer a reliable picture of

investment patterns in the area, which has remained largely unchanged

after the 2008 economic crisis as confirmed by the high correlation with

IMF 2012 data.

Table 2.1 shows that the majority of Italian foreign operations in the

region are concentrated in EU NMs (46.82% of total operations in the

area according to the IMF; 45.39 in FDi Markets), followed by ACC

countries (15.43% for the IMF; 18.52% in FDi Markets), ENP Southern

(20.48% and 10.62% respectively) and ENP Eastern (2.09 for IMF and

6.37% for FDi Markets). Furthermore, a notable share of greenfield

investment from Italy locates in Russia (15.18% in IMF and 19.11% in

FDi Markets). The table suggests that FDi Markets is under-estimating

the share of investments in the ENP Southern countries (ENP-S): indeed,

the dataset looks at the number of new investment projects, and not at

their financial value. The difference between the two measures suggests

that Italian investments in the ENP-S (as will be confirmed by the

interviews) tend to be relatively more capital intensive than in the eastern

countries (ENP-E). Table 2.1 also highlights the importance of Russia as

a destination: it is the single most attractive country in the area under

analysis and, as such, it is an important benchmark for the assessment

of alternative investment locations in the area. Other very relevant

60

locations for Italian foreign operations are Romania, Bulgaria and Poland

in the EU NMs area, with shares equal to 11.2%, 9.65% and 7.92%

respectively. Ukraine in the ENP-E area (4.25%) and Tunisia in the ENP-

S (3.28%) represent the main regional destinations. With respect to the

ACC countries, Italian operations appear more evenly distributed among

regional actors, with an important role played not only by Turkey (4.4%)

and Serbia (4.05%), but also by countries such as Albania (3.47%) and

Croatia (3.28%).

[Table 2.1 here]

Table 2.2 shows Italian foreign investment in the area by business

activity (only available from FDI markets). Following Nielsen (2008) in

classifying activities in core and support business functions, it becomes

apparent that 48.45% of Italian foreign operations in the area involve

‘core business functions’, while 51.53% can be defined as ‘support

activities’. Core functions are strongly dominated by investment in

manufacturing activities (42.47% of total), suggesting that most Italian

MNEs target the area for their ‘production’ activities. With respect to

support functions, investments are dominated by ‘marketing, sales and

after sales servicing’ (32.23%) and ‘administrative and management

functions (13.12%)’. Within the former category, investments are strongly

concentrated in ‘retail’ activities (23.36%) and ‘sales, marketing and

support’ (8.49%), whereas the ‘business services’ sub-category (12.93%)

dominates the latter. The functional classification of the investments

suggests that Italian MNEs are attracted in the area by two fundamental

forces: low-cost production sites (manufacturing investments) and large

and growing markets (sales-related investments).

[Table 2.2 here]

61

Table 2.3 reports Italian MNEs investment projects by broad sector of

activity. The large majority of investment is concentrated in the industrial

manufacturing sector (67.95%), while services represent a smaller share

(26.45%). The majority of manufacturing investments is concentrated in

medium-low technology sectors (47.3%, with textiles accounting for

26.64% of the total) but there is also a relevant share of operations

carried out in high-medium technology sectors (20.66%). In the service

industries, investment in high knowledge-intensive services (16.6%) is

higher than low knowledge-intensive services (9.85%) and it is mostly

dominated by financial services (13.71%). The sectoral analysis suggests

that while business functions are polarised around two key activities, a

broader variety of sectors are involved in the internationalisation process

of Italian investors in the area.

[Table 2.3 here]

This preliminary descriptive evidence on the geography of Italian

investments in the area reflects the more general trends highlighted in

the existing literature. Technological change and the process of EU

integration have favoured a process of structural re-organisation of

Italian foreign investments in traditional sectors such as textiles and

footwear, with the search for new investment targets and international

value chain networks (Amighini and Rabellotti, 2006; Carabelli et al.

2009; Dunford, 2006). EU NMs and NCs have benefitted from this

reorganization of production, receiving a relevant share of Italian

‘production’ and ‘sales’ investments. Italian ‘production’ investments

have been pushed by the strong labour-intensive specialization of the

Italian industrial base confronted with increasing domestic labour-costs

and reduced profit margins in the absence of the competitive

devaluations of the Italian Lira typical of the 1980s and early 1990s

(Resmini, 2000). Conversely, ‘sales’ investments reflect the increasing

62

pressure for access to new (often less sophisticated) markets for Italian

products and services. On a European scale, it has been suggested that

ENP countries strongly benefit from EU foreign investment, which carry

more advanced technological knowledge and managerial practices

(Monastiriotis and Borrell, 2013). This geography of foreign investment is

also reflected in the nature of the trade flows between the EU and NMs

and NCs (Boschma and Capone, 2013; Petrakos et al., 2013; Pinna,

2013), with the latter specializing in less technologically advanced

labour-intensive goods.

The quantitative analysis will explore these processes in a systematic

way making it possible to identify the investments drivers after

controlling for sectoral and functional factors.

2.3 Quantitative analysis

2.3.1 Empirical model and data

In line with existing empirical literature on the location choices of

foreign firms (e.g. Schmidheiny and Brülhart, 2011), a Poisson regression

model is adopted to investigate the relationship between a set of country-

level attributes and the location decisions of 518 Italian greenfield

investment in the region in the period 2003-200814. The number of

investments attracted by each country is modelled as a function of a

number of national characteristics that can be referred back to two key

investment motives discussed below – market-seeking and efficiency and

resource-seeking motives – after controlling for general rule-of-law

conditions and geographical and institutional proximity.

The following equation is estimated:

14

2003 is first year covered by the FDi Markets database. 2008 is the last year not affected by the economic

crisis. Post-economic crisis data are still not available/sufficiently reliable in the FDi Markets database. The

comparison with 2012 IMF investment data has confirmed that FDi Markets data offer a reliable picture of

the geography of Italian investments in the area.

63

𝐼𝑡𝑎 𝑖𝑛𝑣𝑒𝑠𝑡𝑖𝑡 = 𝛼 + 𝛽1𝑚𝑎𝑟𝑘𝑒𝑡 𝑠𝑖𝑧𝑒𝑖𝑡−1 + 𝛽2𝑔𝑜𝑣. 𝑐𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛𝑖𝑡 + 𝛽3𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒𝑖𝐼

+ 𝛽4𝑒𝑥𝑝𝑜𝑟𝑡𝑠𝑖𝑡 + 𝛽5𝑛𝑎𝑡. 𝑟𝑒𝑠𝑜𝑢𝑟𝑐𝑒𝑠𝑖𝑡 + 𝛽6𝑐𝑜𝑛𝑡𝑟𝑜𝑙 𝑜𝑓 𝑐𝑜𝑟𝑟𝑢𝑝𝑡𝑖𝑜𝑛𝑖𝑡−1

+ 𝛽7𝑟𝑢𝑙𝑒 𝑜𝑓 𝑙𝑎𝑤𝑖𝑡−1 + 𝛽8𝑒𝑑𝑢𝑐𝑎𝑡𝑖𝑜𝑛𝑖𝑡 + 𝛽9𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑤𝑎𝑔𝑒𝑖𝑡

+ 𝛽10𝑎𝑔𝑔𝑙𝑜𝑚𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑖𝑡 + 𝛽11𝐼𝑡𝑎𝑙𝑖𝑎𝑛 𝑝𝑟𝑒𝑠𝑒𝑛𝑐𝑒𝑖𝑡

+ 𝛽12𝐸𝑈 𝑚𝑒𝑚𝑏𝑒𝑟𝑠ℎ𝑖𝑝𝑖𝑡 + 𝛽13𝑐𝑜𝑙𝑜𝑛𝑦𝑖 + 𝛿 + 휀𝑖𝑡

where the dependent variable Ita investit is the count of Italian

investment in recipient country i in year t. The explanatory variables are

explained in what follows.

Market-seeking

Market sizeit-1 is the log of National GDP at constant prices (US dollars

2005) in country i with a one-year lag, built on United Nations data. This

is meant to capture the effect played by the internal demand on the

choice of Italian MNEs to locate in recipient countries. There is wide

acknowledgement in the empirical literature that this is a relevant pull

factor for FDI and MNEs strategies (Wheeler and Mody, 1992; Chen and

Moore, 2010).

Government consumptionit stands for general government final

consumption expenditure as a share of GDP in country i and year t. This

represents a proxy for the propensity of the government to incur in

public spending and it might represent a relevant demand factor for

MNEs, although a larger government role is frequently associated to

inefficiencies and rent-seeking (Shleifer and Vishny, 1999). This measure

is taken from World Development Indicators.

Agglomerationit represents the role of agglomeration economies in

attracting foreign investment and it is measured by the share of urban

population in country i and year t, as reported in World Development

Indicators. There are good reasons to believe that more agglomerated

areas are more attractive for foreign investors due to virtuous cycles of

64

externalities (Guimarães et al., 2000). However, considering the

characteristics of Italian MNEs activities in the area, that are strongly

skewed towards Medium-Low technology manufacturing, we might also

expect that these operations are located far from cities to avoid

congestion costs.

Efficiency- and resource-seeking

Average wageit is indirectly measured by means of log per capita GDP

in county i and year t, calculated on data on GDP and population

provided by the World Bank. Data on wages for most countries in the

area are not available or not homogeneous. Existing empirical evidence

on FDI in Central and Eastern European countries suggest that MNEs

tend to locate in these areas for the large supply of cheap labour

(Resmini, 2000). This hypothesis seems reasonable in the present

context, also keeping in mind that investment of Italian MNEs is mostly

concentrated in basic activities.

Educationit is meant to capture the average education level in the host

economy i in time t. This is the log of the ratio between secondary school

age population and total population provided by UNESCO. Considering

the wide and particular set of recipient countries under analysis, this is

the only available measure for plausibly catching an education effect. The

empirical evidence points out that FDI are attracted by locations

endowed with higher human capital (Noorbakhsh et al., 2001; Crescenzi

et al., 2013). Nevertheless, considering that Italian MNEs tend to invest

in manufacturing and retail as well as Medium-Low technology

manufacturing, as Tables 2.1, 2.2 and 2.3 show, we might also expect

that they do not look for relevant human capital in the area.

Natural resourcesit indicates total rents from natural resources as a

share of GDP in country i and year t. The literature has evidenced the

existence of foreign operations from MNEs aimed at exploiting host

65

national natural resources (Asiedu, 2006). This is relevant to test here

considering the set of countries under analysis, which includes large oil

and natural gas producers. This measure is taken from World

Development Indicators.

National Framework Conditions

Control of corruptionit-1 and Rule of lawit-1 are proxy variables for

quality of the national institutional environment in host country i in year

t-1, based on World Governance Indicators. These are aggregate

indicators of different aspects of governance and countries’ institutional

context ranging from 2.5 to -2.5 with higher values associated with more

effective control of corruption and rule of law, respectively. Existing

empirical evidence on the role of institutional factors in determining FDI

and MNEs location behaviour tend to suggest that foreign investors

search for stable and reliable institutional settings to locate their

operations (Altomonte, 2000; Phelps and Waley 2004; Rabbiosi and

Santangelo 2014)

Degree of Integration/Institutional Proximity

Exportsit stands for the value of exports of goods and services as a

share of GDP in country i and year t. We expect a positive correlation of

Italian MNEs location decisions and the importance of exports in host

nations as a sign that MNEs interact with recipient countries also

through trade: in fact, they might locate operations in recipient countries

and re-export goods and services, suggesting an export-platform

rationale of foreign investment (Ekholm et al., 2007). This measure is

based on World Development Indicators.

Italian presenceit, is a stock variable generated on the basis of

previous investment in the same destination country i by nationality (i.e.

other Italian investment). This is to detect any pattern in the decisions of

66

Italian MNEs that my follow national lines. This measure is constructed

with data from FDi Market.

EU membershipit and colonyi are dummy variables that capture

specific characteristics of host countries in term of integration or political

ties (Phelps, 1997). These are provided by CEPII. The former indicates

whether country i is an EU member in year t, as membership to the

Union provides countries with privileged economic and political links

with Italy. The latter indicates whether country i had a past colonial

relationship with Italy.

Geographical Proximity

DistanceiI refers to the geographical distance between host country i

and Italy I, as provided by CEPII. The literature has emphasized the

importance of geographical distance in affecting trade and FDI via

transaction, management and communication costs, arguing that most

proximate locations are generally preferred (e.g. Silva and Tenreyro,

2004).

Finally, δ represents country-year dummies and εit is a random error

term.

2.4 Results and discussion

Table 2.4 shows the results for the estimation of the Poisson

regression model. The regression diagnostics confirm the robustness of

the results and the goodness-of-fit of the model. Column 1 includes all

investments drivers: proxies for market-seeking, efficiency and resource-

seeking, national institutions, degree of integration and institutional and

geographical proximity. In columns 2 and 3 additional controls for degree

of integration/institutional proximity are included: the pre-existing stock

of Italian investments and EU membership together with a control for the

colonial past of the country.

67

Market-seeking factors exert a significant influence on the attraction

of Italian foreign operations in the countries of the area: ceteris paribus

countries with larger internal markets are more likely to be chosen by

Italian investors. In addition, as will be supported by the interviews in

the qualitative section, not only private demand exerts a crucial role for

investments in the area but also public procurement remains central in a

number of sectors and fields of activity: the intensity of government

consumption is a positive and strongly significant predictor for the

presence of foreign operations in a country of the area. The evidence on

the role of both ‘private’ and ‘public/government-led’ demand is robust to

the inclusion of additional controls for the degree of integration/

institutional proximity between the various countries and Italy (columns

2 and 3). What becomes negative and statistically significant after

controlling for the pre-existing links between Italy and the destination

country (as proxied by the pre-existing Italian presence) is the degree of

concentration of the population in urban areas (‘Agglomeration’).

Countries with large densely populated metropolitan areas are – ceteris

paribus – less attractive to MNE investments. This suggests that size of

the national market is a very relevant ‘attraction’ force but its

concentration in large urban areas might rapidly increase congestion

costs (in a context of still un-developed basic infrastructure) discouraging

foreign investments.

The high sensitivity of foreign investments to cost factors and

efficiency motives is confirmed by the negative and strongly significant

impact of average wage levels: high wages discourage investments. The

negative impact of higher wages is not mitigated by higher average skill-

levels. On the contrary, countries with a larger share of secondary

educated people tend to attract – ceteris paribus – less foreign

investments. The coefficient of the ‘Education’ proxy is always negative

and becomes significant in column 2, after controlling for the stock of

pre-existing investments. Once other Italian MNEs have invested in the

68

country – facilitating the upgrading of local suppliers and the provision of

key standardised skills – the overall level of education of the population

discourages new investments. This aspect will be further investigated

with the case study analysis. Finally, the presence of natural resources

exerts a positive and highly significant impact on foreign investments in

all specifications of the model. Resource-seeking motives are still an

important part of the story when considering foreign investments in the

area.

When it comes to the general national ‘framework conditions’ for

foreign investments in the area, ‘control of corruption’ and ‘rule of law’ -

identified by the exiting literature and international organisations as the

key obstacles for foreign investment take off in the region – are positive

and significant predictors for new investments. Countries with more

effective corruption control systems seem to be more attractive to Italian

investments (positive and significant coefficient in column 1). However,

once the stock of pre-existing Italian investments is accounted for, the

more general ‘rule of law’ becomes a positive and significant attractor of

investments, while the specific control of corruption turns out

insignificant.

The final set of regressors control for the degree of economic

integration and institutional proximity between sending and receiving

country. Pre-existing trade flows positively influence subsequent

greenfield investments (column 1) but the direct presence of previous

Italian investments is far more important, making the trade coefficient

almost non-significant. The results highlight a significant path-

dependent aspect in Italian MNEs location behaviour (that will be

confirmed by the case studies), with new investment replicating past

location choices in order to benefit from existing formal and informal

local networks and suppliers linkages. As far as the role of EU

membership is concerned, the regression analysis does not detect any

effect on investments. It is very likely that the most of this effect has been

69

anticipated in the 1990s and early 2000s, as the literature has

highlighted (e.g. Resmini, 2000).

[Table 2.4 here]

2.5 Qualitative analysis

The overall picture of the drivers of Italian investments in the area

and their location strategies developed with the quantitative analysis

needs to be complemented with more in-depth qualitative analysis of

specific case studies of Italian Multinationals with multiple investments

in the EU-15 (the core of the EU) and in the countries of the area under

analysis. Two major Italian MNEs fulfilling these criteria have been

selected for the case studies: Finmeccanica and Saipem. A short

presentation of these companies and their activities in the area will be

followed by the analysis of the interviews15 with key executives in both

firms. A copy of the guidelines/questionnaire used for the semi-

structured interviews with the executives is included in Appendix B.

2.5.1 MNEs profiles

Finmeccanica

Finmeccanica is a major Italian corporate group active in seven high-

technology sectors including Helicopters, Defence and Security

Electronics, Aeronautics, Space, Defence Systems, Energy and

Transportation. As a holding company, Finmeccanica owns 9

enterprises16 operating in these sectors and it also participates into 8

15

Interviews with executives were conducted at the company Head Quarters on April 2, 2013 and May 31,

2013 (Finmeccanica, Rome); June 3, 2013 (Finmeccanica, London); 8 April, 2013 (Saipem, Milan). 16

AgustaWestland, DRS Technologies, Selex ES, Alenia Aermacchi, Oto Melara, WASS, Ansaldo Breda,

Ansaldo STS, BredaMenarinibus.

70

joint ventures17 through its controlled companies. According to the 2013

Finmeccanica Group Profile, it is Italy’s leading industrial company in

high-technology activities and ranks amongst the top ten global players

in Aerospace, Defence and Security. As emerged in the interviews to

executives, 30.2% of Finmeccanica is owned by the Italian Treasury,

which is the largest shareholder of the group. This implies a strong

connection between corporate strategies and the international relations

between Italy and third countries. This is a very relevant feature of this

corporate group, which operates in highly sensitive sectors for Italian

strategic interests.

The international presence of Finmeccanica has strongly increased in

recent years: it employs about 67,000 people in 230 industrial and

technical sites and in 322 commercial and marketing offices in over 50

countries. In terms of sales, Finmeccanica sells its products in nearly

150 nations. From an organizational point of view, it is headquartered in

Italy and has a relevant industrial and commercial presence particularly

in four markets: Italy, UK, USA and Poland. As far as its economic

performance is concerned, revenues in 2012 have reached 17.2 billion

Euros, of which 32% is attributed to Defence and Security Electronics,

24% to Helicopters and 17% to Aeronautics.

As highlighted in the interviews with executives, Finmeccanica is a

large and very complex corporate group, in terms of typology of sectors

and customers, since it has strong ties to both civil and military actors.

This implies highly diversified commercial strategies and approaches

across geography according to the political, institutional and business

profiles of the recipient countries.

Saipem

Saipem is a large multinational company and one of the main world-

17

NHIndustries, ATR, Eurofighter GmbH, SuperJet International, Telespazio, Thales Alenia Space,

MBDA, Ansaldo Energia.

71

wide contractors in the oil & gas industry. It operates mainly in energy-

related activities in remote areas and deep-water, and it is considered a

world leader in the provision of engineering, procurement, project

management and construction services. Saipem’s core business is design

and execution of large-scale offshore and onshore projects with relevant

technological competencies in terms of gas monetization and heavy oil

exploitation18.

In terms of ownership structure, Saipem is part of the ENI (Ente

Nazionale Idrocarduri) group that currently owns approximately 43% of

Saipem. From an organisational standpoint Saipem is organized in two

Business Units: Engineering & Construction and Drilling.

As emphasized during the interview with executives, Saipem is a

global contractor with strong local presence in several European

countries (with key strategic subsidiaries in France, UK, and in new

member states such as Croatia and Romania), but also in emerging areas

such as West Africa, North Africa, Central Asia, Middle East, and South

East Asia. More recently the company has pursued the vigorous

development of production sites in Saudi Arabia and Indonesia, as well

as engineering and project management centres in Algeria, Azerbaijan,

the United Arab Emirates (UAE) and Canada.

A relevant feature of Saipem is that it operates through a highly

decentralized organizational structure in order to take advantage of local

strengths and respond to location-specific needs and sustainability

issues. The company invests substantially in local facilities, ranging from

engineering centres and support yards (for maintenance and storage of

construction equipment) to fully-fledged fabrication yards, where sections

of major projects are assembled for onshore field construction or offshore

18

‘Gas monetisation’ is the development of different typologies of gas from ‘natural resources’ into ‘final

products’ ready for the international markets. This process implies the transformation of the product so as

to match specific modes of transport (e.g. liquid gas transported via dedicated pipelines). Similar challenges

apply to ‘heavy oil exploitation’: heavy crude oil requires prior transformation in order to flow to

production wells. These operations and processes require high technological competences.

72

installation. It also contributes to local employment as a way to enriching

the diversity of Saipem workforce and to recruiting young talents from

around the world.

2.5.2 Analysis of the interviews with executives

The interviews with key executives in both Finmeccanica and Saipem

suggest that market-seeking and resource-seeking investment dominates

the strategies of these two Italian MNEs in the area of interest. These

companies, although being substantially different in terms of sector of

activity, internal organisation and objectives, offer interesting and

illustrative examples of location strategies and modalities of crucially

important MNEs from the same country of origin in the EU-15 towards

EU NMS and NCs.

Mode of Entry

While the quantitative analysis can only look at greenfield

investments (for which systematic data are available) the interviews made

it possible to shed some light on alternative modes of entry of MNEs into

the local markets. Executives in Finmeccanica highlighted in their

interviews that trade connections act as an initial link, but partnerships

with local firms are crucially important to enter new markets. Alliances,

joint ventures, partnerships and M&As are all components of a

diversified strategy to establish a presence in the local markets with new

subsidiaries as the very final step (e.g. in the case of Poland by means of

a key acquisition). Very similar approaches were highlighted by

executives in Saipem. Subsidiaries are used in more sophisticated

relational-intensive contexts in the EU-15 (UK and France), and where

wider markets are expected to be served by means of stable regional

hubs in the NMs (Croatia and Romania). Conversely, in ENP-S and ENP-

E countries partnerships and joint-ventures with local firms are

73

considered the key modes of entry into the local economies (e.g

Azerbaijan or Egypt). The establishment of local offices normally follows

the formation of partnerships in key countries (e.g. Libya with

approximately 100 employees, or Algeria with more than 500) as part of a

gradual expansion strategy in the foreign market.

Market-seeking operations

Regression results suggest that the presence of Italian MNEs in EU

NMs and NCs is higly influenced by the size of national markets.

Moreover, the analysis provides indication that government consumption

is also important as a pull factor for Italian investment. Interviews with

Finmeccanica’s executives reveal that a large share of its operations in

the countries under analysis responds to market-seeking motives.

However, the interviews offer a more nuanced picture of this type of

investment driver.

When looking at investments in NMs, Finmeccanica interviewees

stressed the importance of the acquisition of the Polish firm PZL-Świdnik

in 2010 via its fully-owned sister company AgustaWestland. This

acquisition followed a 20-year long Finmeccanica presence in Poland

through several of its fully-owned companies. Therefore, Finmeccanica

had developed connections and direct experience of the Polish market

during two decades before entering the national market with a direct

acquisition. Before the latter, PZL-Świdnik was already a supplier of

AgustaWestland for several components of helicopters (e.g. fuselage) and,

at the time of the acquisition, around 60% of the activity in PZL-Świdnik

was connected to Finmeccanica. However, according to the interviewees,

the objective of the acquisition was not the in-sourcing of part of the

production chain, but rather a step in a wider strategy aimed at gaining

a strong and more stable presence not only in the Polish market but also

in other Central and Eastern European countries (CEECs) leveraging

Poland as a regional hub. In fact, as far as the Defence sector is

74

concerned, Poland has made substantial investments in the last years

and it represents the main market in the CEECs area. According to

figures of the European Defence Agency, the Defence expenditure of

Poland has increased by 41.3% between 2005 and 2011, reaching €6,557

million in 2011, and it is followed by that of Czech Republic which

stands at only €1,843 million. Also in relative terms, the Defence

expenditure of Poland in 2011 had the largest weight on national GDP

among CEECs, amounting at 1.77%. As compared to the Defence

expenditure of the EU-15 countries, Poland ranks immediately after the

main ‘old’ members: the UK, Germany, France, Italy, Spain and the

Netherlands. Therefore, there are strong indications that the presence of

Finmeccanica in Poland is connected to market-seeking strategies in

response to both private and government-related demand. In this

respect, the preferred mode of entry has entailed the acquisition of a pre-

existing domestic firm, in line with the strategies of most MNEs aiming at

accessing CEECs markets since the later 1990s (Uhlenbruck, 2004).

With respect to NCs, Finmeccanica has a remarkable interest for local

markets in Turkey, Russia and several Northern African countries, such

as Libya, Egypt and Algeria. Expansion in all these countries needs a

constant institutional support of both the Italian and the host

governments, given the strategic national defence importance of some of

Finmeccanica’s products. However, within the complex set of

institutional and political relationships, the selection of the target

countries for Finmeccanica investment is largely driven by market size

considerations and in particular by the importance in the Defence

market. This is especially the case for Finmeccanica-owned firms in

Turkey and Russia, all with a strong commercial orientation towards the

local market.

Market-seeking motives have a very different nature for Saipem given

the specific nature of its goods and services (i.e. engineering,

procurement, project management and construction services). For

75

Saipem – as discussed below – location strategies are closely linked to

the location of natural resources that attract its products and services to

particular locations. However, this demand is often anticipated and

matched by means of appropriately tailored products thanks to constant

interactions established with the key potential customers. These complex

network of contacts and linkages takes place through the subsidiaries

located in London and (to a lesser extent) in the regional hubs in NMs in

Croatia and Romania. Large representative offices in Algeria (ENP-S) and

Azerbaijan (ENP-E) pursue similar – although more peripheral and lower-

level – functions of ‘anticipation and matching of potential demand’.

Efficiency and Resource-seeking operations

From the interviews with Finmeccanica executives it clearly emerged

that the key driver for the selection of Poland as a key hub in the NMs

was the abundant supply of high quality engineers. Given the

significantly lower average wages in Poland vis á vis the other major

locations of Finmeccanica (Italy, UK and USA), the conjugation of market

(discussed above) and efficiency-seeking motives is immediately

apparent. Conversely, the technology and competence gap with the NCs

seems to make it impossible to leverage local human capital in any

significant form. Access to natural resources does not seem to play a

particular role for Finmeccanica given the global and versatile nature of

its value chain.

Conversely, Saipem interviewees suggested that the main rationale for

the location behaviour of their company is linked to the presence of oil

and gas resources and their markets. The time horizon of Saipem

operations in a certain country tends to be more long-term the more

important the location is in terms of energy markets. In the set of

countries under analysis, Saipem has different strategies for different

locations according to their relative importance in terms of resource

endowments. Therefore, Saipem mostly operates in places such as the

76

Russian Federation, Algeria, Libya, Egypt, and Azerbaijan as well as

other locations including Morocco and Tunisia. Hence, as interviewees

pointed out, the main motivation behind the location strategies of Saipem

is not attached to traditional considerations such as efficiency- or purely

market-oriented investment, but it is entirely dependent on the presence

of natural resources. Once operations in a location are established,

Saipem aims at a long-lasting presence, given that natural resources are

immobile. Therefore, labour cost, fiscal incentives, local demand or other

determinant factors for operations in other sectors tend not to be the

primary concern of the location strategy of Saipem in the area

investigated, although they might have a complementary impact. Indeed,

over 75% of total employment in Saipem around the world is represented

by personnel from developing countries where natural resources are

located, suggesting that efficiency-seeking motivations remain important

for the Italian MNE.

National Framework Conditions, Degree of Integration/Institutional and

Geographical Proximity

In line with official policy documents by the European Commission

(2013) and with the results of the quantitative analysis, interviewees at

both Finmeccanica and Saipem agree on the importance of rule of law

and stable and reliable institutions for their operations in the countries

of the region. Highly convergent are also the views of executives in both

MNEs on the very limited influence of geographical proximity for their

location strategies. Both companies highlighted the ‘global’ search for

investments opportunities that is rarely constrained by spatial distance

considerations, although one of the Saipem interviewees highlighted

geographical proximity as an additional factor justifying the selection of

Croatia for one of their subsidiaries.

77

What remains remarkably distinctive in the strategies of both MNEs

is their approach to the ‘development’ of institutional proximity with their

target countries.

A noticeable example of the complex interaction between market-

seeking motives and institutional factors (i.e. the importance of bilateral

inter-governmental relations and agreements) comes from the case of

Finmeccanica in Egypt, where some of the companies owned by

Finmeccanica have experienced a rapid growth in the last few years.

Egypt is a strategic country in the region of Middle East and North Africa

(MENA), with strong political ties with the US. As mentioned in the profile

section, Finmeccanica is also a US ‘domestic’ group by virtue of its

acquisition of the US-based DRS Technologies in 2008. Furthermore, a

number of other controlled or owned companies have strong interests in

the US market. Therefore, Finmeccanica could benefit synergistically

from the strong role played by the US in Egypt and, at the same time,

from the bilateral agreements between Italy and Egypt to operate in this

country.

Saipem has instead adopted a completely different strategy to develop

relationships and integration with its host countries, centred on the

importance of local actors in its activities. Saipem interviewees revealed

that the success of the presence of Saipem in a country is directly

connected to the intensity of interactions with local social and

institutional actors, highlighting the importance of these resources for

the final product. This strategy is based on a trust-building process with

local agents through partnerships, sub-contracting practices and

training of local workforce, leading to the development of a local network

of collaborations that supports corporate activities and objectives.

Successful operations require a certain degree of embeddedness in local

contexts to gain some competitive advantage and secure a long-term

presence in a relevant location.

78

This clearly recalls what has been recently suggested by scholars in

terms of network relationships between MNEs and agents within the

local context (e.g. Crescenzi et al. 2013; McCann and Mudambi, 2005;

Meyer et al. 2011; Iammarino and McCann, 2013), where MNEs embed

their practices in local contexts through their foreign affiliates according

to both corporate objectives and social, economic and institutional

features existing in the specific local environments. Furthermore,

training and employing local workers allows foreign affiliates to generate

and take advantage of new local competitive advantage (Cantwell, 2009;

Phelps and Waley 2004) as well as incorporating local profiles and

competences in MNEs activities and objectives. Following this line of

argument and balancing it with efficiency-seeking considerations,

Saipem’s strategy is to maximize the employment of local personnel.

Indeed, over 75% of total employment in Saipem around the world is

represented by personnel from developing countries where natural

resources are located. The maximization of what the company defines as

“local content” of the activities carried out in foreign markets is one of the

main features of Saipem’s business philosophy. The “local content”

strategy is aimed at providing considerable social benefits to the host

country, in terms of investments, employment, development of

subcontractors and other factors.

Table 2.5 summarizes the key evidence emerging from the case

studies analysis presenting the material in a comparable fashion with the

quantitative regression analysis.

[Table 2.5 here]

2.6 Conclusions

79

This paper analysed the location strategies of Italian Multinationals in

EU NMs and NCs by means of a mixed-methods approach that allowed

us to gain a comprehensive picture of both host location and firm-level

characteristics, which jointly determine MNE choices and strategies. The

regression analysis assessed the relative importance of alternative

country-level features as drivers of location choices, whilst the in-depth

case studies focused on two of the largest Italian MNEs – Finmeccanica

and Saipem – providing relevant insights and complementing the

econometric investigation.

The quantitative and qualitative analyses offer a clear and convergent

picture of the Italian MNE behaviour in the area. However, the case

studies highlighted also significant sectoral and functional differences

across the two firms that would have otherwise remained ‘hidden’ in the

idiosyncratic components of the regression.

The overall results show that market-seeking strategies are still

predominant in driving foreign investments in the EU NMs and NCs.

Both private and government-related demand exerts a very relevant

influence. However, the predominantly low-medium tech production

investments that dominate capital flows between Italy and the area tend

to be discouraged by congestion costs: increasing urbanisation has a

negative impact on investments. The high sensitivity of MNEs to cost

factors (efficiency-seeking) is confirmed by the strong attractive power of

low wages and natural resources; the quality of the general business

environment and the rule of law are, as expected, key facilitating factors

for foreign operations.

If some ‘common’ factors can be generalised from both the

quantitative and the qualitative analyses, the ways in which MNEs enter

the local markets and develop new institutional and functional proximity

with the local economy tend to remain highly diversified. Multinationals’

strategies are influenced by their sector of activity, organisational

structure, strategic management of the value chains and business

80

culture. In the case of Finmeccanica inter-governmental networks and

bilateral international agreements are leveraged to enter local markets

and develop the necessary integration with the target economies. As far

as Saipem is concerned, institutional assimilation with local markets is

developed by means of special arrangements such as local training

initiatives and employment of local workforce (‘local content’), and place-

specific sustainable activities.

In this context the European Neighbourhood Policy (ENP), by

strengthening the links between the EU and its neighbourhood in

institutional, political, social and economic terms, can possibly facilitate

the development of the ‘framework conditions’ needed for EU MNEs

investments in the area. More direct interaction with the European Union

can also facilitate institutional reforms and the pro-investment change in

the individual countries of the area. However, the results presented in

this paper suggest that substantial technological upgrading is still

necessary in order to attract more sophisticated functions and reduce the

current emphasis on purely market seeking investments. In this context,

policies supporting human capital and training (and re-training) of the

local labour force might play a very relevant role.

A note of caution in interpreting these results is needed, as the

different methodologies here implemented can offer only a partial view of

the complexity of MNE strategies. In fact, while the quantitative analysis

provides a picture of the location attributes that drive MNE choices and

the qualitative analysis offers a focus on MNE diversity, generalising

these findings to other contexts can be a misleading exercise. More

research is certainly needed to explore the interaction between location

advantages and MNE heterogeneity in determining FDI decisions for

other samples of countries or regions within countries.

81

Table 2.1: Italian new foreign operations in the EU NMs and NCs

Country Number of New Investment Projects

(2003-2008)*

% Outward Direct Investment Positions

(USD, Millions) 2012**

%

EU New Member States (NMs)

Bulgaria 50 9.65 1015.19 1.46

Czech Republic 15 2.9 1986.65 2.86

Estonia 2 0.39 63.69 0.09

Hungary 29 5.6 2683.77 3.87

Latvia 9 1.74 31.22 0.04

Lithuania 2 0.39 0.08 0.00

Malta 1 0.19 693.60 1.00

Poland 41 7.92 15757.23 22.70

Romania 58 11.2 4749.54 6.84

Slovakia 22 4.25 3887.00 5.60

Slovenia 6 1.16 1634.90 2.36

Subtotal 235 45.39 32502.85 46.82

EU Accession and Candidate Countries (ACC)

Albania 18 3.47 1491.64 2.15

Bosnia and H. 11 2.12 231.80 0.33

Croatia 17 3.28 1063.57 1.53

Macedonia 2 0.39 175.83 0.25

Montenegro 4 0.77 239.12 0.34

Serbia 21 4.05 1074.12 1.55

Turkey 23 4.44 6435.62 9.27

Subtotal 96 18.52 10711.70 15.43

ENP Southern Countries (ENP-S)

Algeria 6 1.16 5889.20 8.48

Egypt 10 1.93 5723.42 8.24

Israel 3 0.58 447.40 0.64

Lebanon 5 0.97 56.11 0.08

Libya 5 0.97 278.38 0.40

Morocco 8 1.54 403.55 0.58

Syria 1 0.19 421.96 0.61

Tunisia 17 3.28 997.21 1.44

Subtotal 55 10.62 14217.22 20.48

ENP Eastern Countries (ENP-E)

Armenia 1 0.19 186.77 0.27

Azerbaijan 4 0.77 175.60 0.25

Belarus 1 0.19 48.81 0.07

Georgia 2 0.39 39.20 0.06

Moldova 3 0.58 122.57 0.18

Ukraine 22 4.25 879.26 1.27

Subtotal 33 6.37 1452.21 2.09

Russia 99 19.11 10536.55 15.18

Total 518 100 69420.53 100.00

* Source: FDi Markets data; **Source: IMF data

82

Table 2.2: Italian new foreign operations in the EU NMs and NCs by business

activity

Business Activity n %

CORE BUSINESS FUNCTIONS 251 48.45

Construction 27 5.21

Manufacturing 220 42.47

Other 4 0.77

SUPPORT BUSINESS FUNCTIONS 267 51.54

Distribution and Logistics 28 5.41

Marketing, sales and after sales servicing 167 32.23

Retail 121 23.36

Sales, Marketing & Support 44 8.49

Other 2 0.38

ICT Services 0 0

Administrative and management functions 68 13.12

Business Services 67 12.93

Other 1 0.19

Engineering and related technical services 2 0.39

R&D 2 0.39

Total 518 100

83

Table 2.3: Italian new foreign operations in the EU NMS and NCs by sector

Sector n %

MANUFACTURING 352 67.95

High-Medium Technology 107 20.66

Automotive Components 12 2.32

Automotive OEM 20 3.86

Consumer Electronics 17 3.28

Industrial Machinery, Equipment & Tools 20 3.86

Other 38 7.34

Medium-Low Technology 245 47.3

Building & Construction Materials 16 3.09

Consumer Products 16 3.09

Food & Tobacco 18 3.47

Textiles 138 26.64

Other 57 11.00

SERVICES 137 26.45

High Knowledge-Intensive 86 16.6

Financial Services 71 13.71

Other 15 2.9

Low Knowledge-Intensive 51 9.85

Hotels & Tourism 14 2.7

Real Estate 16 3.09

Transportation 15 2.9

Other 6 1.16

PRIMARY 29 5.6

Total 518 100

84

Table 2.4: Poisson regression results

Dep.Var.: Investment count 1 2 3

Market-Seeking

Internal market size t-1 2.776*** 1.873*** 1.703***

0.511 0.561 0.612

Government consumption 0.080*** 0.086*** 0.089***

0.011 0.01 0.01

Agglomeration -0.054 -0.111** -0.100**

0.04 0.0432 0.044

Efficiency- and Resource-Seeking

Average wage -1.651** -3.411*** -3.241***

0.656 0.596 0.63

Education -0.447 -1.029** -1.019**

0.502 0.504 0.493

Natural resources rents 0.037*** 0.017*** 0.016***

0.004 0.003 0.003

National Framework Conditions

Control of corruption t-1 0.519*** 0.149 0.14

0.148 0.154 0.148

Rule of law t-1 0.024 0.814*** 0.833***

0.194 0.164 0.164

Degree of Integration/Institutional Proximity

Exports 0.009** 0.008* 0.008*

0.004 0.004 0.004

Italian presence 0.450*** 0.458***

0.0534 0.054

EU membership -0.044

0.055

Ex-Colony 2.427

2.392

Geographical Proximity

Distance 0.007*** -0.005* -0.005*

0.002 0.003 0.003

Constant -63.0*** -6.3 -3.6

10.43 14.13 14.95

Observations 518 518 518

National dummies Yes Yes Yes

log likelihood -3286 -3068 -3065

pseudo R-squared 0.908 0.914 0.915

Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

85

Table 2.5: Summary Table of Case Studies

SAIPEM FINMECCANICA

NMs ENP NMs ENP

Entry mode

Subsidiary

(Croatia,

Romania)

Partnerships and

representative

offices (e.g.

Algeria,

Azerbaijan)

Acquisition

(Poland)

Joint-Ventures

/Partnerships

Market-Seeking Hubs for wider

regions 0

Government Demand /

Hubs for wider

regions

+

Efficiency- and Resource-Seeking

0 + for Natural Resources

+ for Human Capital

0

National Framework

Conditions + + + +

Degree of

Integration/Institutional Proximity

EU

Local

embeddedness and 'local

content'

EU

Bilateral inter-

governmental agreements

Geographical Proximity

Relevant for the

choice of

Croatia

0 0 0

Source: based on interviews with executives

Legend: + Relevant; 0 neutral/not relevant

86

Appendix B

GUIDELINES/QUESTIONNAIRE FOR IN-DEPTH INTERVIEWS TO

MULTINATIONAL ENTERPRISES

SECTION 1: GENERAL INFORMATION ON CORPORATE GROUP 1.Key facts about your Enterprise (e.g. general info, presentation, industrial

sector, core strategy / aim, main facts and figures, etc.)

2.Enterprise structure (e.g. geographical distribution of functions/activities)

3.Where does your Enterprise have operations in the European Neighbouring Policy (ENP) area?

Northern Africa / Middle East East and Caucasus

□ Morocco □ Ukraine □ Algeria □ Belarus

□ Tunisia □ Moldova □ Libya □ Georgia

□ Egypt □ Armenia □ Syria □ Azerbaijan

□ Lebanon □ Jordan

□ Palestine □ Israel

SECTION 2: LOCATION

(When answering this question please refer to ENP countries as indicated in question 3b)

4. What are the main considerations behind the selection of a location for

investment within the ENPs? (e.g. natural resources, new markets, costs/efficiency, strategic assets/competences, etc.)

5. What are the functions or activities that your Enterprise locates in the ENPs? (Headquarter, R&D, marketing/sales, production, logistic &

distribution, etc.)

6. Your presence in the UK and in the ENP area is part of a larger strategy? How? (e.g. creation of a corporate global network, penetration vs.

consolidation, relations with competitors, relations with partners, customers/suppliers, etc.)

87

7. What are the entry modes of your Enterprise in the ENPs? (e.g. joint venture, M&A, sub-contracting, other business agreements, etc.)

SECTION 3: LOCAL LINKAGES

8. What are the localised social and economics actors your Enterprise

establishes relationships with in the ENPs? (e.g. local firms, other foreign subsidiaries, universities/research centres, trade unions, industry

associations, other organisations, etc.)

9. What is the aim of establishing relationships with local actors in the ENPs? (e.g. suppliers/customers, competitors, technological

collaborations/training/joint research projects, institutional support/bureaucracy, etc.)

10. To what extent relationships with local actors are formalised in the

ENPs? (e.g. formal vs. informal, trust-based/control, permanent vs. temporarily relationships, etc.)

11. Does co-location (in the same subnational region/locality) play a role in

determining what local actors are selected for establishing relationships with in the UK? And in the ENPs?

12. To what extent relationships with local actors in the ENPs contribute to

the innovation activities of your Enterprise? (e.g. what kind of knowledge is transmitted through such relationships? Product/process innovation,

solutions to technical problems, project support, basic vs. advanced knowledge, etc.)

SECTION 4: LOCAL CONTEXT

13. What are the strengths and weaknesses of the ENPs in the long-term

strategy of your Enterprise? Please indicate the importance of the following points from 1 (very weak) to 5 (very strong):

□ Labour cost;

□ Quality of human capital; □ Competition;

□ Political framework; □ Regulation/bureaucracy;

□ Institutional quality; □ Technological/scientific base;

□ Business culture □ Other (please specify)

14. How does your Enterprise reacts to the above mentioned weaknesses in

the ENP context? (e.g. training, lobby, etc.)

88

Chapter 3 – Economic Institutions and

the Location Strategies of European

Multinationals in their Geographical

Neighbourhood

3.1 Introduction

Over the past two decades the European Union (EU) has strongly

intensified economic and political relationships with its geographically

neighbouring countries. Two rounds of enlargement in 2004 and 2007

brought several ex-socialist economies under the aegis of the EU, Croatia

joined in 2013, and more countries are currently candidate to

membership. In addition, the European Neighbourhood Policy (ENP) was

launched in 2004, with the aim of creating a ring of countries across the

Mediterranean and the East of Europe with which the EU could intensify

economic linkages as well as develop peaceful and cooperative

relationships (COM, 2004; Wesselink and Boschma, 2012). The complex

set of connections that the EU has established with a wide range of

actors in the area has gradually enhanced the economic and institutional

integration between the EU itself and its counterparts. While full

economic integration was attained with the New Member States (NMS),

the interactions with candidate countries and ENP countries are still

growing.

In this scenario, Multinational Enterprises (MNEs) from the Old EU-

15 members have had wide and increasing opportunities to expand their

operations within the continent and beyond its immediate borders. The

aim of this paper is to study the location of investments undertaken by

EU-15 MNEs towards a wide set of locations integrated or linked to

89

different extents the Union: NMS, Accession and Candidate Countries as

well as ENP countries and the Russian Federation.19 This is a highly

heterogeneous group of EU members, transition and developing

countries, the latter two groups having in common their geographical

proximity to the EU. This entails a set of privileged relationships with the

Union, ranging from full membership in the case of NMS, accession

treaties, action plans within the ENP framework, and bilateral

agreements in the case of Russia.

In particular the paper aims to analyse the role of economic

institutions in shaping MNE greenfield investment location decisions

once new opportunities and geographical options are made available by

tighter economic integration or more favourable preconditions for foreign

investments as a result of formal agreements. By exploiting the unique

conditions offered by the selected group of countries with varying degrees

of economic integration with the EU and highly heterogeneous

institutional conditions, the paper focuses on three key dimensions of

the recipient economies: (i) regulatory characteristics connected to both

national labour markets and business conditions; (ii) legal aspects

relevant in market transactions, i.e. property rights protection and

degree of contract enforcement; (iii) weight of government intervention in

the host countries’ economies.

The contribution of the paper is threefold. First, it innovatively

combines the literature on institutional conditions with the analysis of

MNEs location strategies by focusing, differently from other existing

works, on economic institutions and their different dimensions. In fact,

although the institutional environment of recipient countries has been

the object of analysis of a number of studies, the great majority of this

19

The countries here considered are 21, namely: (a) NMS: Bulgaria, Czech Republic, Estonia, Hungary,

Latvia, Lithuania, Poland, Romania, Slovakia and Slovenia; (b) Accession and candidate countries:

Albania, Croatia (which joined the EU in July 2013) and Turkey; (c) ENP: Ukraine; Algeria, Egypt, Israel,

Jordan, Morocco and Tunisia; (d) Russian Federation.

90

literature focuses on political, rather than economic, features of the

national institutional setting (e.g. Campos and Kinoshita, 2003). Second,

the paper acknowledges right from the start the high heterogeneity of

MNE behaviour with reference to economic institutions, therefore making

use in the empirical strategy of a random-coefficient Mixed Logit (MXL)

model (still rarely employed in this field of research)20 in order to fully

capture this heterogeneity and its drivers.21 The investigation of the

diversity of MNE preferences is still an underdeveloped area of enquiry,

especially as far as quantitative analyses are concerned, while qualitative

approaches have already started to explore such a dimension (e.g. Phelps

and Wu, 2009). Hence, this work contributes to the ongoing scholarly

debate by empirically testing the nature and magnitude of MNE

preferences with respect to recipient countries’ institutions. In so doing,

the paper also explores how heterogeneous preferences in MNE

localisation strategies vary across different sectors of economic activity

and business functions. Third, notwithstanding the increasing geo-

political and economic importance of the EU ‘neighbourhood’, there is

very limited empirical evidence on the (evolving) position in global

investment networks of this set of countries. Filling this gap is crucially

important for the design of appropriate development policies by the

European Union, as well as for national governments and a number of

international organisations active in the area (e.g. United Nation

Development Programme and the World Bank among others).

The analysis is based on the combination of data on 6,888 greenfield

investment projects undertaken between 2003 and 2008 by MNEs from

EU-15 countries into a set of 21 destination countries, and Fraser

Institute data on their economic institutional conditions. The paper

firstly applies a standard Conditional Logit model in order to maximise

20

See Defever (2006; 2012) and Cheng (2008) for previous modelling of MNEs location choices with

random-coefficient Mixed Logit. 21

This methodology allows to model variation of preferences over location attributes in MNEs strategies.

91

comparability with existing studies and, in a subsequent step, explores

MNEs’ behavioural heterogeneity by means of random-coefficient Mixed

Logit. Although we should refrain from any causal interpretation of the

results, the empirical analysis suggests that economic institutions play a

highly significant role in shaping greenfield investment decisions after

controlling for other economic characteristics of the host economies,

showing significant heterogeneity in MNEs’ preferences over different

institutional settings both by sector and by function of the MNE.

The paper is structured as follows: Section 2 provides an overview of

the relevant literature on MNE location behaviour and on the role of

economic institutions in attracting foreign investors, identifying the main

research questions and hypotheses to be tested. Section 3 describes data

and a variable used in the analysis, and provides some descriptive

evidence about the location of European foreign investment in the group

of countries of interest and their institutional conditions. The

methodology is discussed in Section 4, while Section 5 presents the

empirical results. Finally, some concluding remarks and tentative policy

implications are drawn in Section 6.

3.2 MNEs location strategies, host economy

advantages and institutional conditions

3.2.1 MNEs and host economy advantages

The analytical framework for the analysis of MNE location decisions is

Dunning (1977, 1988)’s Ownership-Location-Internalisation (OLI) eclectic

paradigm. The OLI framework implies that the existence of ownership-

specific advantages (O) possessed by some firms may lead to the decision

to internalise (I) activities and to undertake operations in sites endowed

with location-specific advantages (L). Consequently, the combination of

92

(O), (L) and (I) advantages justifies MNEs’ existence and their ability to

maximize their productive efficiency while minimising the impact of

uncertain and imperfect markets on their operations.

However, whilst the interactions between ownership and

internalisation advantages have been extensively investigated (see for

example the seminal work by Buckley and Casson, 1976; Teece, 1977;

Rugman, 1981; Hennart, 1982), the study of location advantages has

suffered from a number of conceptual and empirical constraints, among

which the problematic conceptualisation of space and the severe

restriction in data availability (McCann and Mudambi, 2005; Iammarino

and McCann, 2013).

In the traditional empirical economics literature attention has been

directed to factor endowments in a broad sense, including, among other

location drivers, physical infrastructure (e.g. Coughlin et al., 1991), tax

differentials (e.g. Devereux and Griffith, 1998), policy instruments (Basile

et al., 2008), and labour costs (e.g. Liu et al., 2010). Urban and regional

economics contributions have focused on agglomeration economies,

spatially bounded externalities and the geographical concentration of

economic activity as drivers of MNEs’ location behaviour (e.g. Head et al.

1995; 1999; Guimarães et al., 2000; Crozet et al., 2004; Disdier and

Mayer, 2004; Devereux et al., 2007; Mayer et al. 2010; Hilber and Voicu,

2010; Spies, 2010). Furthermore, empirical studies within the New

Economic Geography have shown that not only MNEs tend to replicate

the location decisions of previous firms with similar attributes, but

agglomeration effects also act through demand linkages (Head and

Mayer, 2004) as well as specialised inputs supply (LaFountain, 2005).

The Economic Geography literature has more recently focussed on

the fragmentation of international activities of MNEs along functional

lines. This stream of research has highlighted that MNE location

93

behaviour and the fragmentation of production processes into different

functions respond to spatial concentration mechanisms (Defever, 2006 &

2012; Strauss-Kahn and Vives, 2009). The concept of Global Value

Chains has been more recently added to this debate with the analysis of

the linkages between MNEs location behaviour along value chains and

the innovative and socio-economic environment of host locations

(Crescenzi et al., 2014). These analyses suggest that MNE location of

different business functions/Global Value Chain stages may follow

different corporate strategies according to the characteristics of the

investor, the location and the specific operation offshored. Besides, the

location choice is influenced by the phase of firms’ life cycle, leading to a

co-evolution of location decisions and accumulation of firms’ capabilities

(Stam, 2007). Entry modes of MNEs into foreign markets are also shaped

by spatial heterogeneity through the interaction between the strength of

local externalities and firms’ competencies (Mariotti et al., 2014).

Technological regimes and systems of innovation conditions have

been extensively analysed in the literature at the intersection between

Economic Geography and International Business (Beugelsdijk and

Mudambi, 2013). The international spatial allocation of MNE activities

tends to be marked by the existence of ‘core and periphery’ patterns

according to the complexity of activities (McCann and Mudambi, 2005),

leading to differences in territorial trajectories and growth dynamics and

to cumulative causation mechanisms (e.g. Cantwell and Iammarino,

1998 & 2001). Since technological development tends to be cumulative in

nature and characterised by elements that are bounded in specific

places, it is suggested that MNEs establish networks for innovation

across locations by tapping into regional profiles of specialisation and

strengthening local technological competencies, thus feeding a regional

hierarchy of centres across and within national boundaries (Cantwell and

Iammarino, 2003). The interactions between regional knowledge bases

94

and MNEs technological strategies are investigated in terms of knowledge

spillovers and externalities, particularly in the European (e.g. Cantwell

and Santangelo, 1999; Cantwell and Piscitello, 2005) and the US context

(Almeida, 1996).

3.2.2 Economic institutions and MNEs investments

The importance of economic institutions for economic performance

and investments is widely acknowledged in the political economy

literature (Knack and Keefer, 1995; Hall and Jones, 1999; Acemoglu and

Robinson, 2005). Economic institutions affect the structure of incentives

in the economy, influencing the stability and predictability of market

(and non-market) transactions. In this sense they play a crucial role in

shaping capital accumulation and (public and private) investments at all

levels (Acemoglu et al., 2005). However, empirical research has primarily

focused on domestic capital formation, with limited attention to the

importance of economic institutions in driving foreign MNE investment

decisions. Institutions influence MNEs’ operations abroad by a) directly

shaping the returns on their investments and the associated risk (direct

effect); b) indirectly impacting upon other key investment drivers such as

human capital and infrastructure (indirect effect) (see Knack and Keefer,

1995).

In particular the existing literature – still rather limited in terms of

geographical coverage – has failed both to agree on the direct importance

of institutional conditions versus other location drivers, and to reach a

clear consensus on what typologies of institutions matter (if at all) for

MNE investment decisions. The seminal contribution by Wheeler and

Mody (1992) – looking at foreign investments of US Multinationals –

combines a number of institutional indicators (including ‘stability of

labour’, ‘red tapes’, ‘quality of the legal system’, etc.) and compares them

with ‘classical’ factor endowment, agglomeration and ‘openness’

indicators. The empirical analysis concludes that US investments abroad

95

are not driven by the institutional environment of the recipient

economies but by other factors only indirectly influenced by institutions:

even though sectoral and geographical heterogeneity turns out to be

significant, factor endowments and openness remain the key investment

drivers.

This evidence has been challenged by a number of subsequent

studies that try to open the institutional ‘black-box’, aiming to

disentangle the relative importance of specific sub-components of the

host institutional environment and its ‘distance’ from that of the MNE’s

home country. Very diverse sets of institutional conditions have been

tested in different studies under the constraint of data availability for

different groups of countries and time periods. Wei (2000) is the first

study to re-open the debate by means of a comprehensive data set on

bilateral FDI flows: his results suggest a negative relationship between

corruption in the host country and FDI. Henisz (2000) looks at the

negative impact of governance costs, while Campos and Kinoshita (2003)

suggest that bureaucracy quality and rule of law are relevant drivers of

FDI. In a similar vein, Globerman and Shapiro (2002) look at both inward

and outward FDI in a large sample of countries, finding a significant and

positive association between MNEs’ investments and a composite

indicator of institutional quality. Meon and Sekkat (2004) investigate the

Middle East and North Africa (MENA) economies suggesting that it is

political risk in general, rather than one particular institutional aspect,

which limits FDI into a given country in the area. Bénassy-Quéré et al.

(2007) – who look at the link between bilateral FDI flows and institutional

quality (captured by means of Fraser Institute indicators as in the

present paper) – conclude that “good institutions almost always increase

the amount of FDI received” (p.780), at the same time stressing the

heterogeneity associated to distance in terms of institutional

96

arrangements between the origin and the destination country of the

investment.

A few complementary studies have looked at MNE location strategies

at the sub-national level: within countries the degree of economic

integration is higher and (formal) institutional arrangements are

generally more homogenous, making it easier to capture the impact of

other aspects of governance quality. Phelps et al. (2003), Phelps (2004),

and Fuller (2005) find evidence of the importance of sub-national

supportive institutions in different areas of the UK. Du et al. (2008)

investigate the location decisions of US MNEs investing in Chinese

provinces over the period 1993-2001 by looking at several indices of

economic institutions. Using a conditional logit model the authors

suggest that US MNE location behaviour reacts positively to stronger

protection of property rights, relatively limited role of government in

business, lower government corruption and more adequate contracting

environment. These elements provide strong incentives to US MNEs to

locate in Chinese provinces.

Another small number of studies have concentrated their attention on

specific economic institutions and MNE behaviour. Three key dimensions

emerge as the core components of economic institutions with a potential

direct impact on the location decisions of foreign investments: regulatory

framework conditions (with reference to both labour and capital

investments, i.e. labour market and business regulations respectively),

the legal environment (property rights and contracts’ enforcement) and

the role of public expenditure in the economy.

Labour market regulation

Existing literature on the relationship between labour market

regulation and foreign investment is scant. Using OECD data, Dewitt et

al. (2003) highlight that unfavourable employment protection differential

97

between destination and origin countries is harmful for investment.

Other studies suggest that more flexible labour markets in recipient

countries are positively correlated to higher inflows of investment from

abroad (Cooke, 1997; Javorcik and Spatareanu, 2005). On the other

hand, locating in a country with a more regulated labour market could

be associated with a firm’s higher productivity: thus, some stages of

production or certain sectors will tend to locate in more regulated labour

markets (Haucap et al., 1997).

Therefore, beyond the conventional belief and weak evidence that

more rigid labour markets represent a cost for foreign investors, it is

possible to argue that countries with different labour market regulations

attract different types of foreign investment. For instance, Lee (2003)

suggests that the existence of labour unions positively affects firms’

greenfield location of new plants in the Korean automotive industry.

Delbecque et al. (2014) – evaluating the impact of labour market

institutions on the location strategies of French MNEs in the OECD

countries – suggest that labour market rigidity might reduce FDI

attractiveness, but the magnitude of the effect is limited when compared

to other investment drivers such as market potential.

Business regulation

The empirical literature on the role of business regulation in general

economic performance has only recently appeared (Djankov et al., 2006).

In this respect, the quality of the business environment is a crucial

determinant of performance since it stimulates investment. Accordingly,

more business-friendly environments can be attractive for MNEs, which

can operate in a context where bureaucratic and administrative costs are

less relevant. Daude and Stein (2007) suggest that the regulatory quality

is the single most important investment driver. Similar conclusions are

reached by Kaditi (2013) looking at South-eastern European countries.

Positive effects of a more deregulated business environment are also

98

suggested by Kaplan et al. (2011): however, the latter study also

highlights that such effects are only temporary and much less important

than conventional wisdom holds. Globerman and Shapiro (2002)

conclude that it is not regulation per se that matters but the

effectiveness of its implementation and enforcement.

Property rights

The role of property rights is widely debated in the existing literature

on economic institutions. Acemoglu et al. (2001) claim that the protection

of property rights plays a crucial role in shaping long-run development

trajectories. First, more secure property rights both encourage

individuals to invest and raise return rates by protecting against

expropriation from the government or powerful groups (Besley, 1995;

Goldstein and Udry, 2008). Secondly, uncertain property rights may

determine costs that individuals have to pay to protect their property.

Thirdly, secure property rights may facilitate gains from trade by

enabling the mobility of assets as factors of production (Besley, 1995). As

a consequence, MNEs may prefer locations where property rights are

better acknowledged and rightfully protected by the legal system. Again

there is no consensus in the empirical literature on the practical

importance of this particular institutional aspect: Bénassy-Quéré et al.

(2007) and Du et al. (2008) find a positive and significant effect, while

Daniele and Marani (2011) suggest that only organised crime works as a

deterrent for foreign investments while there is no effect of other property

rights infringements.

Contract enforcement

The importance of contract enforcement relies on the fact that market

transactions and the general functioning of the economy are more

predictable when economic agents know that contracts will be legally

binding and they can use courts to resolve business disputes. In this

99

respect, Markusen (2001) suggests that MNEs benefit from locations with

strong and reliable contract enforcement since they can credibly commit

to investment. Daude and Stein (2007) find a positive and significant

impact in a large cross section of world economies, Kaditi (2013)

confirms this result for Southern-European countries and Du et al.

(2008) find evidence that better contract enforcement in Chinese regions

attracts US multinationals.

Government Intervention

From a conceptual point of view, a large role of government could lead

to inefficiencies and rent-seeking (Shleifer and Vishny, 1999). Therefore,

MNEs may prefer location where governments play a relatively marginal

role in the economy. For instance, Du et al. (2008) argue that stronger

government intervention in business operations tends to discourage

MNEs from locating in a particular region. Pogrebnyakov and Maitland

(2011) reach similar conclusions looking at the telecommunication

market in Europe and South America. On the other hand, however,

governments often buy products from foreign firms, either directly or

through state-owned enterprises, or purchase goods from domestic firms

that are vertically connected with MNEs’ subsidiaries. In this sense,

larger public sector consumption may be an appealing feature for MNEs

since it increases the size of host countries’ markets.

3.3 Data

3.3.1 MNE Investment

We employ information on individual investment projects undertaken

by MNEs over the period 2003-2008 provided by the FDi Markets-

Financial Times Business database, which includes all cross-border

100

greenfield and brownfield investment.22 Foreign firms’ operations are

identified by Financial Times analysts through a wide variety of sources,

including nearly 9,000 media sources, project data from over 1,000

industry organisations and investment agencies, and data purchased

from market research and publication companies. Furthermore, each

project is cross-referenced across multiple sources and more than 90% of

investment projects are validated with company sources. In addition,

Crescenzi et al. (2014) show that investment decisions captured by this

database are highly correlated with other macro-level data on FDI from

UNCTAD and the World Bank.

Specifically, this paper makes use of investment projects originated in

EU-15 countries and directed towards EU New Member States (NMS) and

European Neighbouring Countries (NCs), the latter being Accession

Countries (ACC), European Neighbourhood Policy (ENP) countries and

the Russian Federation.23 Since the aim of the analysis here is to

investigate MNE location choices, only data on greenfield investment are

considered, since the location of brownfield investment is clearly a

function of greenfield investments undertaken in previous periods:

hence, only greenfield investment are subject to a choice based on

location attributes.

Table 3.1 provides information on new investment projects in 2003-

2008 originating from EU-15 countries in NMS (panel A) and NCs, that is

Balkan and Eastern countries (panel B) and Northern African and Middle

East countries (panel C). It is not surprising that about 62% of EU-15

investors still choose to remain in the EU by selecting a destination

22

In this database joint ventures are tracked only when they lead to new physical operations, whereas

Mergers & Acquisitions as well as other equity investment are not included. Overall, the inclusion in the

dataset is conditional on the fact that investment projects generate new employment or capital investment. 23

Investment from the EU-27 and the whole world towards the same destination countries are also

employed to test the attractiveness of the countries of interest with different samples.

101

among NMS.24 In this area, Romania, Poland and Hungary are the top

three destinations, with about 14.7%, 10.9% and 9.8% of EU-15

investment, respectively. The trend over the 2000s, however, suggests

that the huge attractiveness of NMS reached its peak in anticipation to

the full EU membership and it is now declining, replicating a pattern

rather typical of previous EU enlargements and restructuring. In the

NCs, instead, MNEs’ presence has increased particularly since the mid-

2000s. In terms of cumulative inflows, the most selected destination

outside the European Union is Russia, with a share of 19%. The rest of

the Balkans and the East attracts an additional 10% of EU-15

investment in the area, whilst Northern Africa and Middle East account

for about 8%.

[Table 3.1 here]

3.3.2 Institutional Conditions

A large number of institutional variables are publicly available,

ranging from measures of governance to political indicators.

Nevertheless, as mentioned in previous sections, this paper is primarily

concerned with the notion of economic institutions. The aim is in fact

covering some aspects of national institutional settings that directly

characterise a country’s economic life and affect the degree of

attractiveness towards foreign investment.

In line with other existing studies on foreign investments and

institutions (e.g. Bénassy-Quéré et al. 2007; Delbeque et al. 2011), we

employ data from the Fraser Institute as it provides information for all

countries covered in our analysis. This dataset contains a number of

indicators reflecting several economic dimensions of national

institutional contexts. In particular, we employ the following four

24

Most of NMS entered the EU in 2004, while Romania and Bulgaria joined in 2007.

102

measures of institutional quality: labour market regulation, business

regulation, protection of property rights, and legal enforcement of

contracts. In addition, we use data from the World Bank’s World

Development Indicators (WDI) to include the relevance of government

expenditure in destination countries. With these five indicators we cover

three main areas of the economic-institutional environment: (i) regulatory

aspects (in labour market and business), (ii) legal aspects (property rights

and contract enforcement), and (iii) extent of public intervention in the

economy.

Labour market regulation: our variable for labour market regulation

proxies the flexibility of national labour markets. This is an index

encompassing information on countries’ hiring and firing rules, collective

bargaining, worker dismissal costs, conscription, working hours and

minimum wage. Higher values of the index are associated to more flexible

regulatory settings.

Business regulation: this indicator includes costs associated to

bureaucracy, taxes, bribes and other administrative burdens that may

discourage MNEs from starting a business in a country. As above, this is

an index with higher values reflecting a less regulated environment.

Protection of property rights: we measure property rights protection by

means of an index assuming higher values when property rights are

more protected.

Legal enforcement of contracts: this aspect refers to the capacity and

effectiveness of courts to enforce rules and contracts between parties.

This is measured with an index taking higher values for countries with

better contracting environments.

Government intervention: we employ the percentage of general

government’s final consumption expenditure on GDP, as provided by the

World Bank’s WDI.

103

Table 3.1 above includes information on the characteristics of the

economic institutions of the countries under analysis. Institutional

conditions are heterogeneous across the countries of the EU geographical

vicinity but generally comparable. The NMs show, on average, higher

values of the institutional indicators and generally higher shares of

public expenditure in total GDP when compared to other countries in the

group. The Balkans and the East, in comparison with the NMs, show

lower average values for the economic institution indicators: this group

includes some countries candidate to EU membership, a process that

formally requires gradual institutional convergence towards EU

standards. The final set of countries includes Northern Africa and the

Middle East. In this group average values of the institutional indicators

are upward biased by Israel and Jordan: after excluding these latter two

countries, the average institutional quality of the area is lower than in

the other groups. Overall, the countries covered in the analysis offer an

ample variety of institutional arrangements that is deemed particularly

suitable to test the location behaviour of MNEs.

3.3.3 Other location drivers

The analysis of the link between MNE location choices and economic

institutions requires taking into account other relevant characteristics of

the host economies. In line with the literature on MNE location choices,

this paper employs several control variables that reflect different

potential drivers for the localisation strategies of MNEs.

First, demand is considered as one of the main factors attracting

European investors into foreign markets. Both internal and external

demand is taken into account. Internal demand fundamentally reflects

the market size of the host countries and it is measured through their

own GDP at constant prices, in 2005 US dollars. In line with theory and

104

existing evidence, it is expected that a larger market size will attract more

foreign investors (Wheeler and Mody, 1992; Billington, 1999). External

demand is instead captured by means of a standard market potential

(MP) indicator á la Harris (1954), as customary in the literature. Similar

to the internal market demand, it is expected that market potential is

positively associated with the location strategies of MNEs.

Trade costs are controlled for by employing a measure of geographical

distance between the most populated cities of origin and destination

countries in the sample: intuitively, greater geographical distance is

expected to discourage foreign investors (Bevan and Estrin, 2004;

Kleinert and Toubal, 2010). Furthermore, a dummy variable indicating

national border contiguity between origin and destination countries is

included.

Some characteristics of national labour markets are also controlled

for. The education level of host countries is taken into account by means

of the ratio of secondary school age population to total population.

Notwithstanding the existence of better proxies of human capital at the

national level, this appears to be the only available indicator for the

destination countries in our sample. A positive relationship is expected

between this variable and the location of MNEs. Moreover, the effect of

average wage is indirectly captured through per capita GDP (see Alsan et

al., 2006). Indeed, wage data are rarely available for most destination

countries in the sample and per capita GDP may represent a fair

alternative. A negative relationship is expected between this proxy for

input cost and MNEs location behaviour.

Furthermore, different measures of agglomeration economies are

considered. The percentage of urban population on total population is

included to control for the relative importance of cities in generating

105

externalities (Glaeser et al., 1992; Head et al., 1995). An indicator for the

stock of past foreign investment in location j is constructed. This

measure captures firm-specific agglomeration effects that may derive

from the advantages accruing to an MNE by locating where other MNEs

have previously invested. Hence, the existing stock of investment should

inform whether firms’ past experience drives further location decisions

(Basile et al., 2008). In constructing this variable available information

on brownfield investment is also considered because corporate

expansions signal to a new investor that previous multinational firms

attach additional importance to a specific location. Since the mere count

of investment projects undertaken in previous years does not reveal

much about investors’ behaviour, the analysis takes into consideration

the potential occurrence of a ‘national ownership’ effect in each time

period, which would suggests the existence of patterns in the strategies

of MNEs on the basis of their nationality. Therefore, a stock variable is

generated for each location according to the MNEs’ country of origin: in

line with studies exploring the role of agglomeration externalities, a

positive relationship is expected with the location choice (Wheeler and

Mody, 1992; Barrel and Pain, 1999).

A set of cultural variables includes dummies indicating whether

origin and destination countries share cultural characteristics, thereby

controlling for whether countries speak common official or unofficial

languages, had a common colonizer after 1945, had a colonial

relationship after 1945, and have been a single national entity. These

variables are frequently employed in studies on the internationalisation

decisions of firms (Rauch, 1999; Perez-Villar and Seric, 2014).

Finally, national fixed effects are included to control for any

unobserved factor that operates at the country level and may play a role

in attracting foreign investment.

106

Appendix C provides a description of all variables employed in the

analysis; all are available for years from 2003 to 2008.

3.4 Methodology

3.4.1 Capturing MNEs heterogeneous preferences for economic

institutions: Mixed Logit Models

Following McFadden (1974), the great majority of the empirical

literature on investment location decisions implies that MNE strategies

are fundamentally driven by individual maximization choices. In other

words, it is thought that MNEs select locations on the basis of the

expected utility or profit that each site may yield on the basis of the

characteristics of the host economies. Conditional Logit (CL) models allow

exploring the effect of alternative-specific attributes on the probabilities

that firms select a particular location among the set of alternatives. The

main assumption in the CL is the Independence of Irrelevant Alternatives

(IIA), which implies that the error term εij is independent across

locations.

An extension of the analysis of MNE location behaviour is developed

by implementing a Mixed Logit (MXL) model. This is basically a

generalization of the standard logit and offers the possibility to relax

completely any restriction associated with the IIA. The existing literature

on MNE location choices has rarely employed MXL, despite the

advantages associated to it. Notable exceptions are relatively recent and

include works by Defever (2006; 2012), Cheng (2008) and Basile et al.

(2008). The present analysis implements a random-coefficient derivation

of the MXL, in line with Defever (2006; 2012) and Cheng (2008), with the

aim of analysing whether MNEs have heterogeneous preferences over

location attributes when they strategically select a location for greenfield

107

investment.25 The analysis of the literature has shown that it is

unrealistic to expect unambiguous results. Indeed, this paper aims to

test if the lack of consensus on the role of specific institutional features

of host economies might be explained precisely by the heterogeneity of

MNEs’ preferences over specific institutional attributes. It is plausible

that some MNEs tend to prefer locations with weaker economic

institutions because they aim at bypassing or eluding transparent

market mechanisms when undertaking business operations abroad. For

instance, weaker economic institutions might facilitate rent-seeking or

moral hazard behaviour, the creation of monopolistic positions, or simply

allow capturing a share of host countries’ public resources, through

lobbying, subsidies or less legalized channels, such as corruption. This is

particularly relevant in the case of the present study since the locations

of interest encompass several transition and developing economies that

are characterized by little transparency, weak democratic decision-

making processes as well as strong vested interests that may influence

market mechanisms. To take this into consideration, random coefficients

are attached to variables of economics institutions, while fixed

coefficients are kept for the remaining location drivers.

Accounting for heterogeneity of MNE locations’ characteristics

formally means that the parameter β, associated with an observable

characteristic x of location j, can vary randomly across MNEs. Formally,

the profit equation that each firm maximizes when investing abroad can

be specified as:

(1) 𝜋𝑖𝑗 = 𝛽𝑖′𝑥𝑖𝑗 + 휀𝑖𝑗

25

Basile et al. (2008) adopt an error-component derivation aimed at investigating substitution patterns

among alternative locations.

108

where the vector of parameters β′ for firm i reflects firm’s preference over

observable location attributes x. Thus, in the setting of random-

coefficient MLX parameters β are not fixed as in CL, but they can reveal

MNEs’ taste variation regarding location characteristics. Coefficients vary

across MNEs in the population with distribution density f (β). Following

Train (2003), each MNE knows its own βi (as well as εij) for all alternatives

and select the location that offers higher profit. However, random

coefficients βi remain unobserved and it is only possible to specify a

distribution for them26. By doing this, parameters θ (i.e. mean b and

standard deviation s) of the coefficients βi can be estimated. In this

paper, a normal distribution is specified for random coefficients

associated with economic institutions. Thus, the analysis will inform

whether MNEs exhibit heterogeneous tastes over different economic

institutional settings. The unconditional choice probability to be

estimated takes the following form:

(2) 𝑃𝑖𝑗 = ∫ (𝑒𝛽′𝑥𝑖𝑗

∑ 𝑒𝛽′𝑥𝑖𝑘𝑘

) 𝑓(𝛽|𝜃)𝑑𝛽

This is the MXL probability, which basically consists of a weighted

average of the product of logit equations evaluated at different values of β

and where weights depend on the density f (β | θ) (Train, 2003). As

mentioned, the aim is to estimate parameters θ, which is possible by

means of simulation methods, which allow approximating probabilities

for any given value of parameters θ. Thus, the simulated probability SP is

initially computed as an average probability at different levels of β:

(3) 𝑆𝑃𝑖𝑗 =1

𝑅∑

𝑒𝛽𝑟𝑥𝑖𝑗

∑ 𝑒𝛽𝑟𝑥𝑖𝑘𝑘

𝑅

𝑟=1

26

If the researcher knows βi, this would allow estimating a choice probability similar to CL.

109

where R is the number of draws, or replications. Basically, for

calculating the SPij, the logit equation (2) is computed with each draw r,

and eventually averaged. In the present analysis, R=500. Successively,

SPij is entered into the log-likelihood function to obtain the following

simulated log-likelihood SLL:

(4) 𝑆𝐿𝐿 = ∑ ∑ 𝑦𝑖𝑗𝑙𝑛𝑆𝑃𝑖𝑗

𝐽

𝑗=1

𝐼

𝑖=1

where yij=1 if firm i chooses location j, zero otherwise. Therefore, it is

possible to obtain the Maximum Simulated Likelihood (MSL) estimator

which takes the value of θ that maximizes SLL.

3.5 Empirical Results

All estimations are conducted for EU-15 MNEs investing in European

New Member States, Candidate/Accession, ENP countries and the

Russian Federation. Additionally, estimations on investment from the

EU-27 and the whole world are also run as a benchmark and robustness

check in order to increase the size of the sample of foreign investments.27

3.5.1 Baseline results

Table 3.2 presents the results from CL estimations. Column 1

provides information for the baseline specification. The results suggest

that three out of five indicators of the quality of economic institutions

exhibit a positive and statistically significant relationship with the

location decisions of MNEs: business regulation, government expenditure

and legal enforcement of contracts. Conversely, labour market regulation

and property rights protection are not significant. This specification

27

CL results are qualitatively identical to EU-15 results and are available upon request. The main MXL

results are included in the tables.

110

includes controls for market demand variables, proxies for trade costs

(i.e. geographical distance between origin and destination countries and

a dummy for contiguity), as well as dummies for cultural characteristics.

All controls show the expected sign. Next, in columns 2 and 3, labour

market characteristics such as education level of the population and

average wage are included. Both enter the regression with the expected

signs, although average wage is only weakly significant. Finally, we take

into account agglomeration forces in the last two columns of Table 3.2.

These turn out to be strongly correlated with the location strategies of

MNEs. With the gradual inclusion of all our controls, the relevance of

economic institutions evidenced in column 1 remains unchanged. MNEs

from EU-15 appear to be sensitive to some aspects of the national

economic institutional setting of host countries. More favourable

business regulation, a stronger presence of the state in the economy and

an appropriate contracting environment play a positive role in shaping

the strategic behaviour of MNEs.

[Table 3.2 around here]

Moreover, our more extended specification (column 5) suggests that

internal market size is positively associated with MNE decisions, whereas

market potential becomes non-significant. Similarly, education loses

importance, probably indicating that MNEs from EU-15 delocalize in the

area of interest some business functions for which more basic skills are

needed. Average wage is statistically insignificant. Finally, both measures

of agglomeration are strongly and positively associated with the

dependent variable. This suggests that agglomeration economies are

likely to play a role in attracting MNEs. Similarly, a pattern of localization

that follows national ownership lines emerges. In other words, MNEs

from the same country of origin tend to undertake investment projects in

the same locations.

111

Overall, the CL estimations are in line with the existing literature.

While it is impossible to find any association between MNEs and the

functioning of national labour markets, a less regulated business

environment seems to attract foreign investment. Similarly, with respect

to the legal aspects of economic institutions, different elements play

different roles: the enforcement of contracts is a relevant institutional

aspect for MNEs behaviour suggesting that MNEs are sensitive to the

respect of formal contracts. On the other hand, property rights protection

does not appear to be a driver of location decisions. Finally, the role of

the state is considered as a positive determinant in MNE choices,

presumably because they can take advantage from public intervention in

the economy or because national governments expenditure is also aimed

at consumption. These results suggest that a further investigation of the

heterogeneity of MNE preferences is appropriate: thus, the following

analysis explores the relationship between MNE strategic behaviour and

the economic institutional environment of recipient economies by means

of MXL. This approach makes it also possible to relax the IIA assumption

that treats the substitution of alternative locations rather unrealistically.

3.5.2 Preference heterogeneity

In the MXL estimations heterogeneity is allowed to occur only for

coefficients associated with economic institutions (variables of interest),

while other regressors are kept fixed. Therefore, MXL estimates

coefficient parameters θ, namely means b and standard deviations s, for

variables that are specified to be random. MXL estimation results are

presented in Table 3.3, where the extended specification is run for EU-

15, EU-27 and world MNEs (columns 1, 3, and 5, respectively). As far as

economic institutions are concerned, previous results are largely

confirmed by the estimated means b of coefficients. Regulation is a driver

of MNEs location choices in the context of national business

112

environments, but not in labour markets, although the mean coefficient

for the latter is weakly significant when we consider MNEs from the

whole world. A strong role of government expenditure in neighbouring

countries is perceived as a positive signal by EU-15 MNEs and world

MNEs, while it does not seem to be very relevant for the EU-27 sample

(possibly because some of these investors are from NMS, which may be

relatively more deterred by a large government role in the host economy).

With respect to the national legal framework, a more effective contracting

environment represents an important location determinant for foreign

investment for all MNEs across specifications; as in previous results,

property rights protection exhibits insignificant mean coefficients.

The MXL estimation also provides standard deviations s for the

coefficients of economic institutions, which are specified to vary

randomly. Some of the estimated standard deviations of these coefficients

are statistically significant, suggesting that parameters do vary across

the population of MNEs under analysis. Therefore, standard deviations

can be interpreted as heterogeneity terms and suggest that different

MNEs attach different importance to economic institutions, explaining

the lack of consensus in the existing literature on the importance of some

of their components. Values of b and s are employed in columns 2, 4 and

6 in order to gain insights on the extent of the heterogeneous preferences

of MNE strategies over economic institutions. For instance, in the case of

EU-15 MNEs, the variable for business regulation takes parameters

b=0.475 and s=0.472, such that for 84.4% of the MNE population the

parameter is above zero, while for the 15.6% it is below. In other words,

the large majority of FDI originating in the EU-15 systematically locates

where doing business is characterised by weaker bureaucratic burdens,

while the rest prefers to locate where business is more strongly

regulated. This figure only varies slightly when EU-27 and world MNEs

are considered (80.2% and 76.1%, respectively). More heterogeneous

preferences emerge when we look at parameters related to the protection

113

of property rights. In the case of EU-15 and EU-27 MNEs, estimates

indicate that the population is indeed split into two halves. This balance

between shares of the population with respect to opposite preferences

over property rights protection also explains the insignificance of the

mean coefficient. Finally, as far as the legal enforcement of contracts is

concerned, taste variation over this aspect of economic institutions is far

less pronounced, with most MNEs preferring locations where the

contracting environment is generally certain. Nevertheless, there is a very

small portion of MNEs in the population that decides to locate where

contract enforcement is weaker.

[Table 3.3 here]

Figure 3.1 depicts probability density functions for economic

institutions by employing parameters estimated by MXL: the graphs refer

to those aspects of economic institutions that exhibit significant

heterogeneity terms s.

[Figure 3.1 here]

The heterogeneity of these relationships, particularly regarding

property rights, poses interesting questions on MNEs strategies and their

motives for investing abroad. The source of heterogeneous tastes may be

associated with unobserved factors operating at the firm-level. Therefore,

in order to explore the systematic nature of heterogeneity of preferences

over economic institutions, the MXL models are run by exploiting

information for sectors and business activities of the investment projects

undertaken by MNEs. Data in FDi Markets provides information on these

aspects. On this basis, following the NACE (rev.1.1) classification, we

group sectors into four categories: High-Medium Technology

Manufacturing, Medium-Low Technology Manufacturing, Knowledge-

114

intensive Services (KIS) and Less-knowledge-intensive Services (LKIS).

Similarly, following Crescenzi et al. (2014), we generate three alternative

groups of business functions: Headquarters and innovative activities (HQ

& Inno); Services, sales and logistics (SSL); Production.28 Tables C.2 and

C.3 in Appendix C show the classification of sectors and business

functions, respectively.

Table 3.4 presents the results for MXL estimations of EU-15 location

decisions performed for different sectors, whilst Figure 3.2 plots the

heterogeneous relationships that emerge from such estimations.

[Table 3.4 here]

[Figure 3.2 here]

In columns 1 and 2 of Table 3.4, regressions are run for High-

Medium Technology Manufacturing sectors. The MXL reveals that

regulation of labour markets does not matter for MNE decisions, while

the intervention of the regulator in business has an ambiguous impact:

the majority of MNEs in High-Medium Technology Manufacturing sectors

prefer locations where administrative and bureaucratic aspects of

running a business are less invasive (62.9%), while the rest prefers

countries where businesses are subject to more regulation. Government

expenditure does not play any role in driving MNEs’ behaviour in these

sectors. As far as legal aspects are concerned, MNEs in High-Medium

Technology activities do attach importance to property rights protection

only in 33% of cases. This result might seem surprising since it implies

that a large group of MNEs from EU-15 investing in the area of

neighbouring countries is driven by less robust property rights. However,

28

Differently from Crescenzi et al. (2014), we generate three groups of functions instead of five due to the

low number of observations in certain MNE activities in the countries here considered. Therefore, we

aggregate together certain functions into the same category (e.g. headquarters with innovative activities).

115

this suggests that MNEs operating in High-Medium Tech sectors might

strategically exploit a weaker enforcement of property rights to facilitate

domestic firms’ upgrading and learning (for example in the area of

intellectual property rights, IPRs), while MNEs rely on internal firm-level

protection mechanisms (see Wu 2000 for the case of IPRs in China). With

respect to the legal enforcement of contracts, almost three quarters of

MNEs in High-Medium Technology Manufacturing systematically locate

in places where this aspect of economic institutions is more adequately

protected.

Columns 3 and 4 report results for Medium-Low Technology

Manufacturing. EU-15 MNEs in these activities react more

homogeneously to the quality of national economic institutions than

those in High-Medium Technology Manufacturing sectors. Indeed, a very

large share of MNEs considers strong regulation in business as an

obstacle (87.1%). Also the coefficient on labour market regulation turns

to be marginally significant and positive, suggesting that MNEs in these

activities tend to prefer countries where labour markets are more flexible,

although the statistical relevance of this relationship remains weak. This

finding is perfectly plausible since we are considering EU-15 MNEs that

localise in the EU neighbourhood area operations characterised by a

lower level of sophistication. This is also evidenced by the strongly

negative coefficient associated to our proxy for average wage, signalling

that MNEs in Medium-Low Technology Manufacturing sectors are

motivated by the supply of inexpensive workforce that is generally low-

skilled. With respect to government expenditure, we find that the mean

coefficient b is not significant and the standard deviation s is only weakly

significant. Although these parameters provide a figure of 99.9% of MNEs

driven by more public spending, they should be cautiously interpreted

given their very low statistical significance. MNEs in Medium-Low

Technology Manufacturing activities do not seem to be sensitive to the

116

degree of protection of property rights, while they uniformly attach a

great importance to the possibility to enforce legal contracts.

With respect to control variables, MNEs in High-Medium and

Medium-Low Manufacturing sectors seem to be motivated by different

rationales. Geographical distance and the previous presence of MNEs

from the same origin country are the only common trait in MNEs

strategies. MNEs in High-Medium Technology Manufacturing activities

are substantially attracted by agglomeration forces, suggesting that

MNEs tend to concentrate this kind of activities in urban areas where

they can access a larger supply of labour and competences. Surprisingly,

the education level of the population does not seem to be a relevant

location driver, although our proxy for human capital only takes into

account secondary education, which is probably inadequate for High-

Medium Technology activities. MNEs in Medium-Low Technology

Manufacturing activities, instead, seem to be essentially motivated by

market-seeking and efficiency-seeking rationales, as suggested by the

strongly significant coefficients of market size and average wage. This

finding is in line with the great majority of literature on FDI in transition

economies, which highlight that foreign investors search for new markets

as well as cheap labour in Central and Eastern European countries

(Resmini, 2000).

The right-hand part of Table 3.4 reports results for services: columns

5 and 6 regard KIS, whilst columns 7 and 8 present results for LKIS.

MNEs in KIS tend invariably to take into consideration business

regulation and the legal enforcement of contracts. Again, parameters on

property rights suggest that this element is an ambiguous factor in

determining EU-15 MNE strategies in EU neighbouring countries. As far

as LKIS activities are concerned, results only slightly vary. The

enforcement of contracts turns out to be unimportant for this kind of

services, whilst LKIS seem to positively react to labour markets that are

117

more regulated and to larger government spending. Control variables in

these regressions reveal that KIS benefit of a more educated workforce

and also that location choices globally follow nationality patterns.

Table 3.5 presents the results of MXL performed for different groups

of business functions, while the corresponding Figure 3.3 illustrates the

variation of preferences across them.

[Table 3.5 here]

[Figure 3.3 here]

Columns 1 and 2 in Table 3.5 refer to operations of MNEs in HQ and

Inno activities. Parameters on economic institutions are only significant

with respect to business regulation and property rights protection. The

former exhibits a weak and positive mean coefficient b, while the latter is

still affected by a significant heterogeneity term s that splits the

distribution of preferences into two halves. Our proxy for human capital,

although positive, is not statistically significant, likely due to the fact

that we only consider secondary education. In general, we do not detect

strong drivers of location decisions of MNEs as far as HQ & Inno

activities are concerned. A different picture emerges instead for SSL

activities (columns 3 and 4). A more flexible regulation of business

operations is a positive driver of location decisions for the great majority

of MNEs (83.4%); whilst for the regulation in the labour market almost

60% of MNEs have a positive perception of flexibility, the rest seem to

prefer more regulated frameworks. With respect to legal aspects, nearly

all MNEs find that the legal enforcement of contracts is a crucial element

(92.1%). In addition, SSL are clearly market-seeking motivated, and

MNEs look for a relatively educated and less expensive labour force to

118

employ in these functions. Finally, columns 5 and 6 provide MXL results

for production activities, whose picture appears less complex than for

other business functions. Economic institutions have a very

homogeneous impact and heterogeneity terms are never relevant: more

flexible regulation in business, stronger government spending and

relative easiness in enforcing legal contracts represent attraction forces

for MNE production operations. Moreover, control variables tell that

production activities of EU-15 MNEs are attracted by larger national

markets and tend to exploit local low-skilled and cheap labour.

3.6 Conclusions

In recent years the EU has intensified economic and institutional

integration with its neighbouring countries, though with different

intensity. Some countries have become EU members, some are candidate

for membership, and some others are part of the European Neighbouring

Policy. In this scenario of growing integration, European MNEs have

increased their operations in neighbouring countries through the setting

up of new foreign affiliates.

This paper has examined how recipient countries’ economic

institutions shape the location strategies of EU-15 MNEs in a large set of

developing and transition countries that are geographically close to the

EU. The empirical analysis starts with a standard CL model, as

customary in the literature, and is successively extended to a random-

coefficient MXL, rarely adopted in studies on firms’ location decisions.

Results are robust across specifications with different data samples as

well as across methodologies.

Table 3.6 provides an overall summary of the results on MNE

heterogeneous preferences for economic institutions. In line with the

existing literature our results confirm that the flexibility of the labour

market – one of the top items in ‘traditional’ institutional reform

119

packages – is not systematically associated with the attraction of foreign

investments. On the contrary, favourable business regulation is clearly

an important driver of MNE location choices: when looking at the entire

sample of MNEs large part of the distribution attaches a positive value to

this characteristic. In addition the heterogeneity of preferences seems to

be largely linked to the most sophisticated activities in sectoral (High-

Medium tech sectors) and functional (HQs and Inno) terms.

The analysis of the role of the protection of property rights explains

why the existing literature has so far failed to reach a clear consensus on

its importance: MNEs are indeed strongly divided with reference to this

specific dimension, particularly in the case of the most sophisticated

sectors and functions. Conversely, for the enforcement of contracts the

results highlight clear-cut MNEs’ preferences for more ‘certain’

framework conditions across sectors (with the exception of LKI sectors)

and functions. Finally, the relevance of public expenditure seems to be

limited to production activities, where the government plays an

important role in supporting demand.

[Table 3.6 here]

These results should be interpreted with caution. First, it is important

to bear in mind that the methodology makes it impossible to draw any

causal conclusions. The analysis of location patterns is able to control for

a large number of possible confounding factors but reverse causality is

still a possibility. Second, the time span covered by the analysis is still

limited and the global economic crisis started in 2008, as well as the

dramatic political changes in some of the countries covered in the

analysis, call for extra care in the interpretation of the findings. Third,

even though the innovative use of quantitative methods makes it possible

to shed new light on the heterogeneous behaviour of MNEs with reference

120

to economic institutions, more qualitative work is necessary (and is in

our agenda for future research) in order to explore the firm-specific

determinants of MNEs’ diversified preferences.

Having acknowledged these limitations, our results provide policy

makers with relevant insights to support institutional reform and

institution building initiatives as tools to favour (and complement)

internationalisation processes. The empirical results suggest that some

MNEs prefer locations where specific dimensions of economic institutions

are weaker. This may appear counterintuitive, but indeed there could be

situations in which economic actors may prefer loose economic

institutions in order to gain selective economic rewards. This

institutional subversion phenomenon is particularly documented in the

case of transition economies, where political and economic elites

replicate a system of flawed institutional environments that provide them

with various types of advantage over the rest of the local population

(Helmann, 1998; Helmann et al., 2000). Similarly, weak property rights

allow wealthier foreign actors to benefit from unproductive activities such

as rent-seeking, at the same time maintaining expropriation instruments

over the rest (Sonin, 2003). The subversion of economic institutions is

also intimately associated with within-country inequality, and less secure

property rights and weaker legal systems favour a country’s power

establishment, which aims at perpetuating the mechanisms that allow

the concentration of power and wealth (Glaeser et al., 2003). In this vein,

it is argued that political incumbents support imperfect institutions in

order to maintain their benefits (Glaeser and Shleifer, 2002). On the

basis of these considerations, often made with respect to transition and

developing countries, it can be argued that some MNEs are oriented

towards locations where they can establish influential connections with

political and economic elites, which in turn allow them taking advantage

of institutional poorness by obtaining rents or circumventing market

121

rules. A similar argument is proposed in the management literature:

pervasive government corruption can influence the entry modes of MNEs,

which can find it beneficial to enter new markets via FDI by engaging in

corrupt behaviour (Rodriguez et al., 2005). Again, this may represent one

explanation for the heterogeneity of results associated to the protection of

property rights in particular. However, validating these results and

investigating further the relationship between economic institutions and

MNEs remain an open research field and a crucial challenge for policy

design in a growing number of countries and regions worldwide.

122

Table 3.1: EU-15 investment projects and quality of economic institutions, 2003-2008.

MNEs Investments Quality of Economic Institutions

Host Countries N of investment % investment

Labour market regulation

Business regulation

Protection of property rights

Legal enforcement of

contracts

Government expenditure

A. New Member States

Bulgaria 551 8.00 6.96 5.60 4.09 4.77 17.97

Czech Republic 443 6.43 7.47 5.16 5.72 3.59 21.46

Estonia 142 2.06 5.87 7.37 7.25 6.02 17.58

Hungary 674 9.79 6.84 6.12 6.51 7.06 22.45

Latvia 152 2.21 6.43 6.29 5.88 7.25 18.50

Lithuania 139 2.02 5.45 6.50 5.80 7.35 19.04

Poland 748 10.86 6.52 5.49 4.66 4.27 18.12

Romania 1,012 14.69 5.91 6.54 4.77 5.17 12.19

Slovakia 319 4.63 7.61 5.85 5.98 4.59 18.42

Slovenia 100 1.45 5.44 6.34 6.27 3.93 18.46

Subtotal / Average* 4,280 62.14

6.45* 6.13* 5.69* 5.40* 18.42*

B. Balkans and the East

Albania 38 0.55 5.79 5.67 3.30 5.17 9.31

Croatia 139 2.02 5.65 5.62 4.70 5.40 19.95

Russia 1,315 19.09 6.03 4.73 3.34 7.53 17.38

Turkey 298 4.33 4.09 6.29 5.06 6.16 12.34

Ukraine 263 3.82 6.22 4.08 3.40 5.29 18.18

Subtotal / Average* 2,053 29.81

5.56* 5.28* 3.96* 5.91* 15.43

C. Northern Africa and Middle East

Algeria 105 1.52 4.96 5.62 4.25 4.39 12.43

Egypt 84 1.22 5.01 5.06 5.77 3.41 12.03

Israel 37 0.54 4.84 6.64 6.98 3.46 25.71

Jordan 23 0.33 8.38 6.45 7.18 3.38 22.01

Morocco 203 2.95 3.62 6.09 5.62 4.3 18.31

Tunisia 103 1.50 6.30 6.79 7.00 4.88 16.67

Subtotal /Average* 555 8.06 5.52* 6.11* 6.13* 3.97* 17.86*

Total / Average* 6,888 100 5.97* 5.92* 5.41* 5.11* 17.55*

Source: own elaboration based on FDi Markets – FT Business and Fraser Institute Data

123

Table 3.2: Conditional Logit estimation of EU15 MNEs location behaviour

Dep.Var.: Location choice (1) (2) (3) (4) (5)

Labour Market Regulation 0.018 0.028 0.044 -0.004 -0.010

(0.043) (0.044) (0.045) (0.049) (0.049)

Business Regulation 0.401*** 0.393*** 0.382*** 0.371*** 0.434***

(0.057) (0.057) (0.058) (0.058) (0.058)

Government Expenditure 0.059*** 0.065*** 0.0623*** 0.067*** 0.045***

(0.014) (0.014) (0.014) (0.014) (0.015)

Protection of Property Rights 0.0017 0.012 0.026 0.010 0.005

(0.039) (0.039) (0.040) (0.040) (0.040)

Legal Enforcement of Contracts 0.567*** 0.559*** 0.560*** 0.683*** 0.591***

(0.128) (0.129) (0.127) (0.138) (0.139)

ln Market Size t-1 -0.455 0.352 1.189 0.919 2.441**

(0.781) (0.837) (0.961) (0.974) (0.988)

ln Market Potential t-1 1.728** 2.405*** 2.591*** 2.044** 0.979

(0.860) (0.891) (0.896) (0.911) (0.917)

Distance -0.001*** -0.001*** -0.001*** -0.001*** -0.001***

(0.000) (0.000) (0.000) (0.000) (0.000)

ln Education Level

1.291*** 0.977** 0.487 0.709

(0.470) (0.495) (0.527) (0.530)

ln Average Wage

-1.343* -0.402 -0.963

(0.777) (0.854) (0.860)

Urban Agglomeration

0.149** 0.151***

(0.058) (0.058)

National Ownership

0.003***

(0.001)

Observations 148,783 148,783 148,783 148,783 148,783

Cultural dummies Yes Yes Yes Yes Yes

Geographical contiguity Yes Yes Yes Yes Yes

National dummies Yes Yes Yes Yes Yes

Pseudo R2 0.193 0.194 0.194 0.194 0.196

log likelihood -17084 -17080 -17078 -17075 -17037 Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

124

Table 3.3: Mixed Logit estimation of MNEs location behaviour

(1) (2) (3) (4) (5) (6)

EU15 MNEs EU27 MNEs World MNEs

Dep. Var.: Location Choice θ Value % > 0 Value % > 0 Value % > 0

Labour Market Regulation b 0.007

0.024

0.072*

(0.051)

(0.049)

(0.039)

s 0.015

0.171

0.008

(0.036)

(0.192)

(0.016) Business Regulation b 0.475*** 84.4% 0.522*** 80.2% 0.403*** 76.1%

(0.064)

(0.063)

(0.047)

s 0.472***

0.613***

0.567***

(0.113)

(0.100)

(0.074) Government Expenditure b 0.035**

0.021

0.025**

(0.016)

(0.015)

(0.012)

s 0.001

0.001

0.001

(0.001)

(0.001)

(0.001) Protection of Property

Rights b 0.002 50.4% 0.035 54.4% 0.001

(0.043)

(0.042)

(0.032)

s 0.229**

0.322***

0.133

(0.097)

(0.085)

(0.103) Legal Enforce of Contracts b 0.570*** 98.4% 0.500*** 94.7% 0.467*** 89.3%

(0.148)

(0.138)

(0.110)

s 0.265***

0.309***

0.376***

(0.097)

(0.094)

(0.069)

ln Market Size t-1

1.963*

2.688***

2.148***

(1.018)

(0.748)

(0.563) Distance

-0.001***

-0.001***

-0.001***

(0.000)

(0.000)

(0.000) ln Market Potential t-1

1.247

1.080

-0.588

(0.977)

(0.885)

(0.680) ln Education Level

0.536

1.184**

0.708*

(0.552)

(0.478)

(0.392) ln Average Wage

-1.490*

-1.997***

-1.662***

(0.887)

(0.729)

(0.576) Urban Agglomeration

0.146**

0.0754*

0.098***

(0.060)

(0.041)

(0.031) National Ownership

0.004***

0.006***

0.006***

(0.001)

(0.001)

(0.001) Observations

148,783

165,724

251,276

N of Cases

6,888

7,709

11,745 Geographical contiguity Yes Yes Yes

Cultural dummies

Yes

Yes

Yes National dummies

Yes

Yes

Yes

log likelihood -17030 -18974 -29437

Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

125

Figure 3.1: Probability Density Functions for economic institutions exhibiting

significant standard deviation in Table 3

84.4%15.6%

mean=0.475

sd=0.472

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

EU15 MNEs

Business Regulation

80.2%19.8%

mean=0.522

sd=0.613

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

EU27 MNEs

Business Regulation

76.1%23.9%

mean=0.403

sd=0.567

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

World MNEs

Business Regulation

50.4%49.6%

mean=0.002

sd=0.229

3sd2sd 1sd0-1sd-2sd-3sd

Probability Density Function

EU15 MNEs

Protection of Property Rights

54.4%45.6%

mean=0.035

sd=0.322

3sd2sd 1sd0-1sd-2sd-3sd

Probability Density Function

EU27 MNEs

Protection of Property Rights

98.4%1.6%

mean=0.570

sd=0.265

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

EU15 MNEs

Legal Enforcement of Contracts

94.7%5.3%

mean=0.5

sd=0.309

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

EU27 MNEs

Legal Enforcement of Contracts

89.3%10.7%

mean=0.467

sd=0.376

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

World MNEs

Legal Enforcement of Contracts

126

Table 3.4: MXL estimation of EU-15 MNEs location behaviour by sector

(1) (2) (3) (4) (5) (6) (7) (8)

Manufacturing Services

High-Medium Tech. Medium-Low Tech. Knowledge-intensive Less-knowledge-int.

Dep. Var.: Location Choice θ Value % > 0 Value % > 0 Value % > 0 Value % > 0

Labour Market Regulation b -0.030 0.149* 0.002 -0.246**

(0.128) (0.083) (0.112) (0.123)

s -0.105 0.005 0.013 0.206

(0.688) (0.020) (0.026) (0.244)

Business Regulation b 0.232 62.9% 0.572*** 87.1% 0.383** 0.406***

(0.160) (0.106) (0.157) (0.152)

s 0.707*** 0.507*** 0.310 -0.014

(0.265) (0.145) (0.405) (0.020)

Government Expenditure b -0.013 0.043 99.9% 0.022 0.086**

(0.040) (0.026) (0.034) (0.039)

s -0.016 0.002* 0.008 -0.000

(0.026) (0.001) (0.011) (0.001)

Protection of Prop. Rights b -0.189** 33.0% 0.086 -0.011 49.2% 0.046 55.6%

(0.093) (0.069) (0.099) (0.105)

s 0.423* -0.019 0.528*** 0.333*

(0.217) (0.019) (0.113) (0.178)

Legal Enforc. of b 0.539 72.6% 0.740*** 0.725** 0.095

Contracts (0.381) (0.239) (0.325) (0.318)

s 0.894** 0.229 0.235 -0.004

(0.389) (0.221) (0.234) (0.025)

ln Market Size t-1

-0.648

4.576***

0.910

0.450

(2.518)

(1.242)

(1.742)

(1.814) Distance

-0.001***

-0.001***

-0.001***

-0.001***

(0.000)

(0.000)

(0.000)

(0.000) ln Market Potential t-1

2.338

0.720

3.135

0.717

(2.752)

(1.593)

(1.922)

(2.377) ln Education Level

-1.262

0.286

2.844**

0.101

(1.400)

(0.830)

(1.286)

(1.367)

ln Average Wage

0.593

-3.821***

-0.234

-0.905

(2.172)

(1.289)

(1.799)

(1.764) Urban Agglomeration

0.432***

0.105

-0.029

-0.021

(0.142)

(0.072)

(0.090)

(0.107) National Ownership

0.003***

0.004***

0.004***

0.003***

(0.001)

(0.001)

(0.001)

(0.001)

Observations

31,039

56,795

28,065

27,357 Geographical contiguity

Yes

Yes

Yes

Yes Cultural dummies

Yes

Yes

Yes

Yes National dummies

Yes

Yes

Yes

Yes log likelihood -3497 -6394 -3230 -3039

Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

127

Figure 3.2: Probability Density Functions for economic institutions exhibiting

significant standard deviation in Table 4

62.9%37.1%

mean=0.232

sd=0.707

3sd2sd 1sd0-1sd-2sd-3sd

Probability Density Function

High-Medium Technology Manufacturing

Business Regulation

87.1%12.9%

mean=0.572

sd=0.507

3sd2sd mean 1sd0-1sd-2sd-3sd

Probability Density Function

Medium-Low Technology Manufacturing

Business Regulation

33.0%67.0%

mean=-0.189

sd=0.423

3sd2sd 1sd0-1sd-2sd-3sd

Probability Density Function

High-Medium Technology Manufacturing

Protection of Property Rights

49.2%50.8%

mean=-0.011

sd=0.528

3sd2sd 1sd0-1sd-2sd-3sd

Probability Density Function

Knowledge-intensive Services

Protection of Property Rights

55.6%44.4%

mean=0.046

sd=0.333

3sd2sd 1sd0-1sd-2sd-3sd

Probability Density Function

Less Knowledge-intensive Services

Protection of Property Rights

72.6%27.4%

mean=0.539

sd=0.894

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

High-Medium Technology Manufacturing

Legal Enforcement of Contracts

128

Table 3.5: MXL estimation of EU-15 MNEs location behaviour by business function

(1) (2) (3) (4) (5) (6)

HQ & Inno SSL Production

Dep. Var.: Location

Choice θ Value % > 0 Value % > 0 Value % > 0

Labour Market b -0.003

0.069 58.7% -0.078

Regulation

(0.138)

(0.081)

(0.077)

s 0.011

0.312*

0.037

(0.008)

(0.185)

(0.089)

Business Regulation b 0.328*

0.527*** 83.4% 0.443***

(0.190)

(0.109)

(0.088)

s 0.512

0.541***

0.265

(0.369)

(0.157)

(0.239)

Government

Expenditure b -0.029

0.015

0.083***

(0.041)

(0.025)

(0.024)

s -0.002

0.001

-0.006

(0.003)

(0.002)

(0.005)

Protection of Prop. b -0.015 48.8% 0.071

-0.070

Rights

(0.118)

(0.066)

(0.064)

s 0.550***

-0.097

0.193

(0.138)

(0.249)

(0.159)

Legal Enforce of b -0.027

0.544** 92.1% 0.764***

Contracts

(0.397)

(0.221)

(0.207)

s -0.271

0.386**

0.203

(0.231)

(0.157)

(0.155)

ln Market Size t-1

0.816

4.108***

2.505**

(2.070)

(1.234)

(1.094)

Distance

-0.001***

-0.001***

-0.001***

(0.000)

(0.000)

(0.000)

ln Market Potential t-1

0.794

1.960

-1.596

(2.199)

(1.522)

(1.433)

ln Education Level

1.849

1.839**

-1.458*

(1.559)

(0.767)

(0.880)

ln Average Wage

0.953

-2.382*

-2.790**

(2.117)

(1.219)

(1.153)

Urban Agglomeration

0.037

0.099

0.116*

(0.106)

(0.069)

(0.063)

National Ownership

0.003***

0.004***

0.004***

(0.001)

(0.001)

(0.001)

Observations

19,994

64,381

64,408

Geographical contiguity

Yes

Yes

Yes

Cultural dummies

Yes

Yes

Yes

National dummies

Yes

Yes

Yes

log likelihood -2293 -7372 -7204

Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

129

Figure 3.3: Probability Density Functions for economic institutions exhibiting

significant standard deviation in Table 5

58.7%41.3%

mean=0.069

sd=0.312

3sd2sd 1sd0-1sd-2sd-3sd

Probability Density Function

Services, Sales and Logistics

Labour Market Regulation

83.4%16.6%

mean=0.527

sd=0.541

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

Services, Sales and Logistics

Business Regulation

48.8%51.2%

mean=-0.015

sd=0.550

3sd2sd 1sd0-1sd-2sd-3sd

Probability Density Function

Headquarters and Innovation

Protection of Property Rights

92.1%7.9%

mean=0.544

sd=0.386

3sd2sd 1sd0 mean-1sd-2sd-3sd

Probability Density Function

Services, Sales and Logistics

Legal Enforcement of Contracts

130

Table 3.6: Summary Table of the Results on MNEs heterogeneous preferences for Economic Institutions

All MNES

Sectoral Heterogeneity Functional Heterogeneity

Manufacturing Services

High-Medium

tech

Medium-low tech

Knowledge

Intensive

Less Knowledg

e Intensive

HQ & Inno

SSL Production

Regulatory settings

Labour Market Regulation

NO NO NO NO NO NO NO NO

Business Regulation

+*** s***

(84%)

s***

(63%) +*** +** +*** NO

+*** s***

(83%)

+***

Legal Framework

Property Rights

s***

(50%)

-** s* (33%)

NO

s***(49%) NO

s***

(49%) NO NO

Enforcement of Contracts

+*** s***(98%)

s**(73%)

+*** +** NO NO +**

s**(92%) +***

Weight of the Government

Share of Public Spending

+** NO NO NO NO NO NO +***

+/- denotes the sign of the estimated b coefficients in tables 3,4 and 5. Asterisks denote significance as in original tables. Percentages reported in parentheses are %>0 in the preferences distribution. ‘NO’ stands for

‘No significance’

131

Appendix C

Table C.1: Variable definitions and sources

Variable Description Source

Dependent

Location Choice Dummy indicating location choices among 23 destination countries

FDi Markets

Independent

Economic Institutions

Labour Market Regulation

Index (0-10) indicating the flexibility of labour market in location j.

Fraser Institute

Business Regulation

Index (0-10) indicating the administrative and bureaucratic burdens for business in location j.

Fraser Institute

Protection or Property Rights

Index (0-10) indicating the extent to which government protects property rights in location j.

Fraser Institute

Legal Enforcement of Contracts

Index (0-10) indicating the extent to which contracts are enforced by courts in location j.

Fraser Institute

Government expenditure

Percentage of general government final consumption expenditure on GDP in location j.

WDI

Demand

Ln Market Sizet-1 Log of GDP of destination j at time t-1. WDI

Ln Market Potentialt-1

Log of the sum of distance-weighted GDP of all countries c within 1,000km from location j at time t-1,

i for each c≠j.

WDI / CEPII

Trade Costs

Geogr. Distance Physical distance measured in km. CEPII

Geogr. Contiguity Dummy equal to 1 if country of origin r and destination j are contiguous.

CEPII

Labour Market

Ln Education Level Log of the ratio between secondary school age population and total population in location j.

UNESCO

Ln Average Wage Log of per capita GDP in location j. WDI

Agglomeration

Urban Agglomeration

Percentage of urban population on total population. WDI

National

Ownership

Stock of investment in location j from the same

country of origin r of firm i. FDi Markets

Culture

Official Language Dummy equal to 1 if country of origin r and location j share an official common language.

CEPII

Unofficial Language

Dummy equal to 1 if country of origin r and location j share an unofficial common language.

CEPII

Common Colonizer after 1945

Dummy equal to 1 if country of origin r and location j had a common colonizer after 1945.

CEPII

Colonial Link after 1945

Dummy equal to 1 if country of origin r and location j had a colonial tie after 1945.

CEPII

Same Country Dummy equal to 1 if country of origin r and location j have been part of the same country in the past.

CEPII

132

Table C.2: Classification of sectors

Manufacturing

High-Medium Technology Medium-Low Technology

Aerospace Beverages

Automotive components

Building and Construction

Materials

Automotive OEM Consumer Products

Biotechnology Food and Tobacco

Business Machines and Equipment Metals

Ceramic and Glass Minerals

Chemicals Non-Automotive Transport OEM

Consumer Electronics Paper, Printing and Packaging

Electronic Components Plastics

Engines and Turbines Rubber

Industrial Machinery, Equipment and Tools Textiles

Medical Devices Wood Products

Pharmaceuticals

Semiconductors

Services

Knowledge-Intensive Less Knowledge-Intensive

Business Services Hotels and Tourism

Communications Leisure and Entertainment

Financial Services Real Estate

Healthcare Transportation

Software and IT Services Warehousing and Storage

Space and Defence

133

Table C.3: Classification of business functions

Headquarters and innovative activities

Business Services

Headquarters

Design, Development and Testing

Education and Training

Research and Development

Services, Sales and Logistics

Customer Contact Centre

Logistic, Distribution and Transportation

Maintenance and Servicing

Recycling

Retail

Sales, Marketing and Support

Shared Services Centre

Technical Support Centre

Production

Construction

Electricity

Extraction

ICT and Internet Infrastructure

Manufacturing

134

Part II: Selection Patterns in

Cross-border Acquisitions

135

Chapter 4 – Cross-border acquisitions

and patterns of selection: Productivity

vs. profitability

4.1 Introduction

In the last two decades a substantial preference for mergers and

acquisitions (M&A) over greenfield FDI has been frequently observed in

global modes of entry by multinational enterprises (MNEs) (Barba

Navaretti and Venables, 2004; UNCTAD, 2010). This is particularly the

case of FDI among industrialized countries, where market access is often

attained through the acquisition of a pre-existing domestic firm rather

than by building a new establishment.

Yet, academic research has only very recently started to distinguish,

theoretically and empirically, between different modes of FDI (i.e. M&A

vs. greenfield) although their characteristics, causes and implications

differ significantly (Nocke and Yeaple, 2007; 2008). Hence,

understanding what shapes selection in cross-border acquisition choices

of MNEs represents a relevant area of enquiry for its academic novelty as

well as its importance in terms of share of acquisitions in global FDI

volumes. In this respect, this paper explores the importance of two main

alternative factors underpinning MNEs decisions to acquire a specific

target firm: namely, a productivity argument related to accessing foreign

valuable assets possessed by target firms, and a profit consideration

associated with the expansion of corporate business in new foreign

destinations.

Consider, for instance, from Chapter 2, the case of PZL-Świdnik, a

polish manufacturer of helicopters acquired by the Italian conglomerate

136

Finmeccanica through its Anglo-Italian subsidiary AgustaWestland in

2010. According to the Chairman and CEO of Finmeccanica this

acquisition generates strong opportunities for the parental industrial

group because of both the expertise of PZL-Świdnik in producing

aerostructures as well as the access that this specific takeover gives to

new and profitable markets29.

This example not only demonstrates that the productivity and

profitability of target firms are crucial factors that MNEs take into

account when engaging in cross-border takeovers, but it also suggests

that distinguishing between these two elements is not always

straightforward as they can be simultaneously at work.

The empirical study of the selection decisions of MNEs is surprisingly

underdeveloped in the literature, mainly due to the lack of time varying

information on firm ownership. Indirect empirical findings in the

literature on FDI-induced spillovers suggest that MNEs tend to ‘cherry-

pick’ best performing domestic firms (Arnold and Javorcik, 2009;

Ramondo, 2009; Criscuolo and Martin, 2009). Only in most recent years

scholars have started to engage in the empirical investigation of the

selection decisions of MNEs, providing initial evidence supporting target

firms’ productivity as a motivating factor of international takeovers

(Guadalupe et al., 2012; Blonigen et al., 2014).

Building on this theme, this paper assesses the extent to which the

probability faced by domestic firms of being acquired in any given year

relates to their productivity and profitability. Conceptually, these

motivating factors can be ascribed to two traditional hypotheses in the

theory of MNEs, namely asset-seeking and market-seeking behaviour of

global companies. The joint assessment of these hypotheses employing

firm-level data represents a first novelty of this paper, in that past

studies on cross-border acquisitions mainly focus on productivity

29 Finmeccanica Press Release “AgustaWestland acquires helicopters and

aerostructures manufacturer PZLSwidnik”, Rome 18 August 2009.

137

differentials. Nonetheless, as evidenced by the example presented above,

while productive assets and capabilities embedded in existing domestic

firms can be relevant aspects that MNEs take into account in the

selection of an acquisition target, MNEs can also engage in cross-border

takeovers to obtain a significant spot in a specific market. The latter

strategy is in line with the objective of gaining direct access to the

existing and promising business linkages of the acquired firm. Hence,

domestic firms experiencing positive changes in their profits over time

may be plausibly selected for acquisition.

In order to separately analyse the effect of target firm productivity and

that of profitability, we exploit within-firm differences in the probability of

being acquired, similarly to Blonigen et al. (2014), and we additionally

compare acquired firms with those that are never acquired in the study

period in order to alleviate any concern related to sample selection.

Hence, domestic firms experiencing positive changes in their

businesses and profits may be more plausibly selected for acquisition.

This paper is also innovative as we conduct the study on a large

sample of European manufacturing firms, as opposed to previous studies

that only focus on companies in single countries or on industry- and

country-level data. Our panel is drawn from Bureau Van Dijk databases

Orbis and Zephyr and it includes 306,247 potential target firms observed

at multiple points in time over the period 1997-2013. In addition, by

employing time varying ownership information on domestic companies

we are able to observe at what point MNEs acquire domestic firms.

Our main empirical finding is that domestic companies that

experience positive changes in profitability have higher probability than

others of being acquired over the sample period. A within-firm increase of

one standard deviation in profitability as compared to the industry mean

is associated to a 0.8% higher probability of being acquired by a foreign

MNE in the next period. By contrast, within-firm variation in productivity

does not significantly relate to international acquisition decisions,

138

suggesting that MNE selection only occurs on the observable market

performance (i.e. profitability) of domestic firms. These findings are

confirmed also by employing different measures of firm productivity and

profitability. Furthermore, baseline results still hold across a large

number of checks and extensions, indicating that within-firm differences

in profitability are intimately associated to changes in ownership.

Understanding the selection patterns of cross-border takeovers is

highly relevant for public policies in both territorial and industrial

perspectives. In presence of FDI-induced spillovers, in fact, designing

regional and industrial programmes aimed at FDI attraction can be

beneficial for the recipient economy. In addition, acquired firms could

benefit from the enlarged market that being part of a global production

chain entails, with potential positive effects also on domestic employment

and on the local network of suppliers.

The paper is structured as follows: the next section is devoted to a

critical discussion of the literature on international acquisitions and

setting up of hypotheses. Section 3 presents data and the construction of

the dataset. Section 4 explains the empirical setting of the paper and its

differences as compared to previous studies. Results are presented in

Section 5 along with a discussion of the findings associated to several

extensions and robustness checks. Section 6 offers some concluding

remarks as well as considerations for policy.

4.2 Related literature

The notion of cross-border investment is intimately associated with

the conceptualisation of the boundaries of the MNE, thus, encompassing

the idea of a trade-off between integration and outsourcing of activities

overseas. This appears to be a nontrivial choice for the management of a

MNE, faced with the issue of internalisation of a specific operation via

FDI and its governance costs. From a theoretical standpoint, the

139

international trade literature has conceived the internalisation decision

as a transaction-cost problem (Grossman and Helpman, 2002) or as a

response to the issue of incomplete contracts between partner firms

(Antrás, 2003; Antrás and Helpman, 2004 and 2008).

Once a MNE decides to undertake FDI, it can do so mainly by

establishing a new plant (greenfield FDI) or by acquiring an existing

domestic firm. This organisational choice depends upon a number of

elements such as recipient country attributes, industry characteristics

and MNEs features (Nocke and Yeaple, 2007; 2008). While the

determinants of greenfield FDI have received wide empirical attention by

researchers, mainly through analyses of location behaviour, there is still

a substantial lack of systematic evidence on the drivers of selection

decisions of MNEs when they undertake cross-border acquisitions.

Reasonably, cross-border acquisitions, far from being casual business

choices, follow specific paths that spring from the interplay between the

complexity of internalisation strategies of MNEs and the characteristics

of heterogeneous domestic firms. In this respect, a nascent strand of

literature has commenced to explore this area of enquiry shedding light

on a number of factors driving MNE selection decisions. In the remaining

of this section, these recent contributions will be reviewed and discussed.

4.2.1 Acquisitions to access foreign productive assets

The evidence that MNEs expand overseas by acquiring domestic firms

in foreign countries is often interpreted as a corporate strategy aimed at

enhancing MNEs existing capabilities (Caves, 1996). This form of asset-

seeking investment is regarded as an expedient of MNEs to advance their

competitiveness at the global level through the enlargement and

140

deepening of their portfolio of tangible and non-tangible assets30

(Dunning and Lundan, 2008).

An underlying assumption in the logic of asset-seeking investment is

that some firms possess assets that are desirable to other firms,

including pure capital goods, specific technical competencies or

managerial and marketing skills (Iammarino and McCann, 2013). Hence,

acquisition activity can be aimed at accessing these assets, which lead in

turn to the realisation of efficiency gains through the exploitation of

similarities between the acquirer and the target firm. In Jovanovic and

Braguinsky (2004), for instance, better managers tend to buy better

projects and the complementarity between the qualities of their assets

lead to the generation of surplus. Nocke and Yeaple (2008) develop an

equilibrium model to explain greenfield FDI and cross-border takeovers,

arguing that MNEs engage in acquisitions in order to complement own

assets with target firms’ assets. In other words, acquisitions lead MNEs

to purchasing complementary activities overseas that the acquirer

initially lacks. In their model, hence, a mechanism of positive assortative

matching entails that better entrepreneurs purchase better production

facilities, thus generating higher profits. A further motive for engaging in

international acquisitions recalls the resource-based view of the firm and

it contemplates the existence of non-mobile capabilities owned by local

firms (Nocke and Yeaple, 2007). MNEs are thereafter pushed to acquire

domestic firms abroad in order to exploit strategic complementary

capabilities that are not transferable across borders. In line with the

complementarity of assets view, Head and Ries (2008) adopt a gravity

and multi-country analytical framework to study bilateral and

multilateral FDI, suggesting that cross-border acquisitions function as

30

Such a view also provides the cornerstone for evolutionary conceptualizations of MNEs, where

FDI serves as an instrument to define and refine new corporate technological trajectories (e.g.

Cantwell, 1989; Kogut and Zander, 1993).

141

an exploitative mechanism of corporate control of overseas productive

assets.

Recently, a number of contributions investigate more specifically the

incidence of target firms’ assets in motivating cross-border acquisitions.

In analysing Norwegian plant-level data, Balsvik and Haller (2010) argue

that foreign owners tend to acquire domestic firms in order to obtain

efficiency gains from synergies associated to the existence of

complementary resources between MNEs and local companies. The

relevance of assets matching as a triggering factor for cross-border

acquisitions is also corroborated by Guadalupe et al. (2012), who

examine the acquisition decisions of MNEs for a sample of Spanish firms.

In investigating the relationship between foreign ownership and

innovative capacity of newly acquired companies, they argue that

incentives for acquisitions and innovation are strongly interdependent

and, as a consequence, a positive selection in acquisition choices occurs

whenever there is a complementarity between target firms’ productivity

and the amount of innovation. In other words, target firms’ productive

assets complement MNEs investment in innovation upon acquisition and

this conducts to the takeover of most productive domestic firms within

industries. Analogously, Blonigen et al. (2014) inspect the dynamics of

cross-border acquisitions on a panel of French firms focussing on the

synergic role played by the capacity of companies to generate export

networks and time-changing productivity levels of these domestic actors.

Their empirical analysis suggests that valuable assets sought by MNEs

pertain to the antecedent capability of French firms to form export

linkages, which is positively dependent on high initial productivity.

Nonetheless, acquisitions are actually found to occur mostly when firms

are afflicted by a negative productivity shock, which generates a

depressing effect on the price of the same assets. On the other hand,

assets complementarity does not emerge as a compelling factor to explain

international acquisitions in Díez and Spearot (2014). In fact, in testing

142

whether assortative matching acts as a meaningful driver of cross-border

takeovers, these authors do not observe the occurrence of this feature in

the data.

4.2.2 Acquisitions to access foreign markets

Cross-border acquisitions can also be motivated by profitability

considerations made by MNEs as a way to increase their market power,

reduce the competitive pressure within industries and attain a larger

market share. For instance, the limited availability of firm-specific

ownership advantages, such as a superior technology, pushes firms to

merge in an oligopolistic market (Horn and Persson, 2001). In this

framework, low trade costs encourage cross-border acquisitions since

firms can access new foreign markets, while high trade costs intensify

domestic mergers due to reduced home competition. In a similar vein,

Bjorvatn (2004) argue that economic integration increases market

competition, thereby reducing the profit and reservation price of target

firms. This, in turn, would raise the gains associated to international

acquisition activity. Evidence in favour of the positive effect of decreasing

trade costs on cross-border acquisitions is provided by Coeurdacier et al.

(2009), as well as by Breinlich (2008), both emphasising the role of

mergers and acquisitions in the process of industrial restructuring

following economic integration. The incentives to engage in cross-border

mergers in an oligopolistic context are also magnified by the existence of

information asymmetries, which encourage uninformed foreign MNEs to

acquire domestic firms with detailed knowledge about demand in the

local market (Qiu and Zhou, 2006). Market power considerations as

drivers of international acquisitions emerge in Neary (2007), where trade

liberalisation is conducive to cross-border merger waves. In fact, with

increased economic integration more efficient firms tend to acquire

foreign less efficient rivals, thus facilitating specialisation according to

143

countries’ comparative advantage. In this respect, Brakman et al. (2013)

and Feliciano and Lipsey (2015) provide evidence that cross-border

acquisition activity is more concentrated in sectors that are characterised

by a revealed comparative advantage in the country of the acquirer.

Although market power considerations and profitability are posited to

be noteworthy aspects spurring cross-border acquisition activity,

empirical tests employing firm-level data are scarce. Early attempts in

this direction come from the industrial organisation literature on

domestic mergers, where the probability of target companies of being

acquired depends upon their level of profitability among other factors

(e.g. Harris et al., 1982; Ravenscraft and Scherer, 1989). The present

study also aims at testing the relevance of domestic firm profitability in

shaping the patterns of selection associated to cross-border takeovers.

4.2.3 Hypotheses development

Considering all the above, this paper posits that MNE acquisition

choices can be driven by two fundamental and interconnected factors:

target firm productivity and profitability. In this respect, the empirical

part of the present paper aims at testing the following hypotheses.

Productivity hypothesis: MNEs acquire domestic firms that exhibit

larger positive variation in productivity over time, as a strategy to access

valuable and complementary assets.

Profitability hypothesis: MNE acquire domestic firms that exhibit a

larger positive variation in profitability over time, as a strategy to access

new market opportunities and expand their business activities.

Not surprisingly, firm productivity and profitability can be

interconnected as more productive firms are more likely to experience

thriving business conditions. Also, firms that experience more profitable

business can increse their productivity as a result of economies of scale.

In the empirical section of the paper, we aim at testing the above

144

hypotheses bearing in mind that these two firm characteristics are

strongly related from a conceptual point of view.

4.3 Data

4.3.1 Dataset construction

Our sample of European companies is drawn from Bureau van Dijk

cross-country and longitudinal databases Orbis and Zephyr. Orbis

provides firm-level information on accounting and financial items of

companies worldwide from which we construct our measures of

profitability and productivity. Data on M&A operations are contained in

Zephyr, which allows tracking time varying ownership information of

firms. The two datasets can be easily matched via common company

identifiers. Previous research employing these sources of data is well

established and it includes recent works on international taxation (Voget,

2011), productivity (Maffini and Mokkas, 2011; Gal, 2013) and bank

lending (Giannetti and Ongena, 2012) among others. In our empirical

analysis, we consider acquisitions occurred from 1997 to 2013 in 14

European countries, that is, EU-15 countries31 with the exception of

Luxembourg, for which no relevant manufacturing firm is observed. For

our purpose, a cross-border acquisition is defined as a transaction

involving a foreign company acquiring a stake of a previously

domestically-owned firm. Thus, the acquirer is a foreign-owned company

and the target is a domestic firm. We therefore exclude from this

definition certain types of operations, such as (i) wholly domestic

transactions where both the acquirer and target are domestic companies;

(ii) domestic firms acquiring foreign affiliates located in the acquirer

country; (iii) transactions involving two foreign entities, such as a foreign

31

These are the so-called ‘Old’ EU member countries: Austria, Belgium, Denmark, Finland, France,

Germany, Greece, Ireland, Italy, Netherlands, Portugal, Spain, Sweden and the UK,

145

affiliate acquiring another foreign affiliate in a third country (iv)

operations resulting in increased stakes of ownership: the latter may

include, for instance, an MNE that already owns a certain percentage of a

domestic firm as a result of a previous cross-border takeover, and

successively engaging in a new acquiring operation to increase its control

over the domestic firm.

Also, we exclude mergers from our empirical analysis, since these

transactions involves a merging of companies on a one-to-one share

swap for shares in the new company32. Hence, while in an acquisition a

firm buys and subsumes another firm, a merger represents a transaction

where two or more firms decide to create a new company. Similarly, we

also exclude other forms of transactions such as joint ventures,

Institutional Buy-Outs (IBOs), Management Buy-Outs (MBOs) and share

buyback operations. Unfortunately, not all cross-border acquisitions in

Zephyr could be matched with company information in Orbis, due to

different issues such as missing observations for the acquired companies

in Orbis before the transactions and some missing identifiers. Other

acquisitions from Zephyr, instead, could not be used in the empirical

analysis because the target firms are not registered in Orbis.

After carefully considering all the above, the dataset includes 458 cross-

border acquisitions. Table 1 reports the number of cross-border

acquisitions and the number of firms by country based on the discussion

above. The sample consists of 306,247 firms observed at multiple points

in time over the period 1997-2013, for a total of 1,177,895 observations.

This results in an unbalanced panel of firms located across 14 countries.

As noticed by other studies using the Orbis database (e.g. Maffini and

Mokkas, 2011), the share of firms in the sample is skewed towards

certain countries such as Italy, Spain and France and this depends to a

large extent upon the availability of key variables across countries. As far

32

Definition from Zephyr user guide online.

146

as cross-border acquisitions are concerned, the largest economies in

Europe, that is, Germany, France, Italy and the UK, account for almost

69% of the total number of transactions. Including Spain in this group

raises this percentage to about 79%.

A restricted sample is generated encompassing only those firms that

are acquired by an MNE over the years 1997-2013. In other words,

domestic companies that are never acquired in the sample period are

excluded from this second dataset. The sample size is then reduced to

268 firms acquired over the sample period and 759 observations. This

reduction in the number of acquired domestic firms is due to the

methodology adopted for the construction of variables, as explained in

the next section.

[Table 1 here]

4.3.2 Variables construction

In order to test our hypotheses relative to the two different drivers of

cross-border acquisitions, two proxy variables for productivity and

profitability of domestic firms are required. We follow the financial

literature in defining the profitability of firms as the ratio between

earnings before interest and taxes (EBIT) and fixed assets (Dewenter and

Malatesta, 2001; Campa and Kedia, 2002; Cornett et al., 2008). EBIT is

calculated in Orbis as the difference between gross profit of a firm, the

total cost of goods sold and other operating expenses. Since EBIT is

calculated before taxes and interest expenses, it provides a good measure

of the ability of companies to make profits. As mentioned, EBIT is divided

by fixed assets as a measure of firm total capital. Finally, the variable is

normalised by its industry mean and logged, as follows:

147

𝑝𝑟𝑜𝑓𝑖𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦𝑖𝑡−1 = 𝑙𝑛(

𝑒𝑏𝑖𝑡𝑎𝑠𝑠𝑒𝑡𝑠)

𝑖𝑡−1

1𝑁

∑ (𝑒𝑏𝑖𝑡

𝑎𝑠𝑠𝑒𝑡𝑠)𝑠𝑡−1

𝑛𝑠=1

(1)

where i denotes the firm, t stands for time and s indicates the NACE 4-

digits manufacturing sector33. Industry means are only calculated by

year and sector in this measure, even if they could be also computed by

country. For instance, a domestic firm in a specific country can be

acquired because it is particularly profitable in its home country.

However, considering the high level of economic integration of EU

countries and the tight trade linkages across Europe, our preferred

measure of profitability is normalised on a wider industry mean than the

country level. Nonetheless, results are checked against alternative

measures of profitability, also taking into account such a national

dimension, are contemplated. First, the effects of depreciation and

amortization of assets are excluded from firm earnings by substituting

EBIT with a measure of earnings before interest, taxes, depreciation and

amortization (EBITDA). The latter can be relevant for capital-intensive

firms and sectors where the depreciation of capital can strongly depress

earnings as measured by EBIT. Second, two additional measures of

profitability are generated by replicating EBIT- and EBITDA-based

variables normalised on industry means calculated by individual country

for the reasons discussed above.

As far as labour productivity is concerned, following to Guadalupe et

al. (2012), this is intended as the ratio between value added and

employment, normalised by industry mean, as follows:

33

The sample includes 292 different NACE 4-digits manufacturing sectors.

148

𝑙𝑎𝑏𝑜𝑢𝑟 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑣𝑖𝑡𝑦𝑖𝑡−1 = 𝑙𝑛

(𝑣𝑎𝑙𝑢𝑒 𝑎𝑑𝑑𝑒𝑑𝑒𝑚𝑝𝑙𝑜𝑦𝑚𝑒𝑛𝑡

)𝑖𝑡−1

1𝑁

∑ (𝑣𝑎𝑙𝑢𝑒 𝑎𝑑𝑑𝑒𝑑𝑒𝑚𝑝𝑙𝑜𝑦𝑚𝑒𝑛𝑡

)𝑠𝑡−1

𝑛𝑠=1

(2)

where i denotes the firm, t stands for time and s indicates the NACE 4-

digits manufacturing sector. A different proxy for labour productivity is

also computed by replacing value added with turnover. Furthermore, as

in the case of profitability, the two measures of labour productivity are

re-computed on industry means by country.

Although TFP may be a better proxy for firm productivity than labour

productivity, the calculation of TFP in Orbis is likely to lower the number

of observations and potentially decrease the number of cross-border

acquisitions that could be used in the empirical analysis, due to the high

requirements for TFP calculation in terms of data. Furthermore, the

decrease in the number of firms may be biased towards companies that

provide a wider range of data, and that are plausibly larger and more

productive than others34. Regardless of these potential limitations,

however, we test the robustness of our results also with respect to two

TFP measures. Table 2 provides the correlation matrix between the

various measures of profitability and labour productivity described in

this section (panel A). Interestingly, profitability and labour productivity

do not exhibit high correlation coefficients. In particular, the correlation

between our preferred measures (PR1 and LP1) is only 0.13. Panel B of

Table 2 provides the correlation coefficient between profitability and

labour productivity in the dataset restricted to domestic companies that

are acquired at some point during the sample period.

34

Gal (2013) shows a very high correlation between TFP and labour productivity

(calculated as value added-employment ratio) using Orbis data.

149

Finally, we consider employment and fixed assets as control variables

for firm size and capital availability35. Summary statistics are described

in Table 3. Interestingly, the mean values of both profitability and labour

productivity are higher in the restricted sample than in the full sample.

Considering that the former only includes domestic firms that are

acquired by an MNE at a certain time over the sample period, such a

difference in mean values may be suggestive of the fact that firms that

are going to become foreign affiliates tend to be more profitable and

productive than the others. Similarly, these firms also tend to be larger

as well as having larger capital endowments.

[Table 2 and 3 here]

4.4 Empirical strategy

In this section we introduce the empirical setting adopted to evaluate

the relevance of the two main hypothesised factors motivating cross-

border acquisitions, that is, the search for productive assets and market

considerations. By employing different measures for labour productivity

and firm profitability, we model the selection decision of MNEs as the

linear probability that domestic firms can be acquired at any time during

the sample period. Covariates are included with a one-year lag in order to

avoid that target characteristics are influenced by foreign ownership. In

this respect, Fich et al. (2011) argue that an M&A negotiation period

typically lasts between 31 and 163 days from the initiation date.

Furthermore, also previous empirical contributions adopt a single year

lag to model acquisition decisions (e.g. Guadalupe et al., 2012; Blonigen

et al., 2014).

35

Similar to profitability and labour productivity, these control variables are normalised by yearly

industry means and take a logarithmic form: 𝑋𝑖𝑡−1 = 𝑙𝑛(𝑥)𝑖𝑡−1

1

𝑁∑ (𝑥)𝑠𝑡−1

𝑛𝑠=1

.

150

Thus, the probability y that a domestic firm i operating in industry s

is acquired in a given year t is estimated as:

𝑦𝑖𝑡 = 𝛽𝑃𝑅𝑖𝑡−1 + 𝛾𝐿𝑃𝑖𝑡−1 + 𝜗𝑋𝑖𝑡−1 + 𝛿𝑡 + 𝜆𝑠𝑡 + 𝜔𝑐𝑡 + 𝜑𝑖 + 𝑢𝑖𝑡 (3)

where PR stands for firm profitability, LP indicates labour productivity, X

is a vector of time-varying firm-level controls, δ is a set of time dummies,

λ includes industry trends, ω represents a set of country-year dummy

variables, φ incorporates firm fixed effects and u is an idiosyncratic error

term. Although the only control variables included in X are firm lagged

employment and fixed assets, we are confident that incorporating fixed

effects at the firm-level will account for any independent, target-specific

and time invariant acquisition determinant that is omitted in the model.

These, for instance, can include managerial quality and practices,

company structure, reputation effects and all sorts of unobserved time-

constant factors operating within the firm that can attract takeovers or

can be correlated with the capacity to generate earnings or employing

assets efficiently.

We also control for specific influences that can affect cross-border

acquisition decisions across years by including time dummies. In fact, it

is well documented that aggregate M&A occur in waves (Andrade et al.,

2001) and such a cyclical nature of corporate business can affect the

probability of firms to be acquired in a given year. Moreover, waves of

mergers tend to be clustered within industries as a result of the exposure

of firms to technological, regulatory and economic shocks that alter the

structure of specific industries (Mitchell and Mulherin, 1996). Hence,

industry trends are included in our empirical model to account for any

time variant industry-specific disturbance that can affect domestic firms’

characteristics as well as the strategic decision of MNEs to incur in a

cross-border takeover and select a specific target. A third important

151

dimension of the non-uniform distribution of acquisitions controlled for

is the geographical dimension. In fact, the clustering of acquisitions in

specific countries is striking in our data, as evidenced in Table 1.

Although firm fixed effects include the location of targets and we do not

have data on firms that move in space across time, we generate a set of

country-year dummies that allow controlling for the concentration of

cross-border takeovers in specific destinations over time. The relevance

of national boundaries and geography for the occurrence of international

acquisitions tends to be associated with the performance of national

stock markets, which are more likely to affect a country as a whole

rather than a specific industry (Erel et al., 2012).

With respect to existing empirical strategies modelling the selection

decisions of MNEs, we combine the above-mentioned aspects in a novel

way. For instance, while accounting for industry trends and time fixed

effects, Guadalupe et al. (2012) explore within-industry differences in

probability of international acquisitions, not controlling for fixed effects

operating at the level of individual firms in their linear probability

specification. Blonigen et al. (2014) extend their baseline logit analysis to

include firm and time fixed effects using a sample that only includes

acquired foreign affiliates. As evidenced in the equation to be estimated

presented above, instead, our empirical strategy combines firm and time

fixed effects with industry trends in a linear specification, employing both

a full dataset including acquired firms as well as those that are never

acquired, and a restricted dataset only containing companies that are

acquired at some point over the sample period. In addition, considering

that the present study focuses on a set of countries rather than a single

country, we also incorporate a term capturing waves of acquisitions that

cluster within national economies. In so doing, we investigate the

relevance of within-firm variation in profitability and labour productivity

in affecting the selection decisions of MNEs that engage in cross-border

takeovers.

152

4.5 Results

This section is structured in six parts, each coinciding to a different

empirical approach employed to test our hypotheses regarding the

selection decisions of MNEs in cross-border takeovers. First, the baseline

analysis concentrates on the relevance of lagged profitability and labour

productivity in driving the choices of MNEs towards certain target firms

rather than others. Second, we scrutinize alternative specifications of the

baseline setting by introducing and testing different measures of firm

profitability and labour productivity. Third, we explore the potential non-

linearity of the selection decision of MNEs as far as the interaction

between profitability and labour productivity is concerned. Fourth, we

compare target profitability and labour productivity across industries

characterised by different technological intensity. Fifth, we adopt a more

stringent definition of cross-border takeovers by re-estimating the linear

probability model on acquisitions of majority stakes as well as completed

takeovers. Finally, we assess the relevance of target characteristics by

restricting the sample to include only domestic firms that are acquired at

some point over the sample period.

4.5.1 Probability of foreign acquisition: baseline estimates

The baseline results for the estimation of the linear probability

equation are provided in Table 4. In columns (1) and (2), lagged measures

of firm profitability and labour productivity are entered in isolation.

[Table 4 here]

This preliminary evidence suggests that, conditional on being

domestically-owned before acquisition, a target firm’s higher ability to

exploit market opportunities and make profits matter for the selection

153

decisions of MNEs. On the other hand labour productivity does not

appear to be a relevant driver of acquisition decisions, suggesting that

takeovers in Europe are not associated with the search for valuable

productive assets. The lack of significance on labour productivity could

be due to the fact that other controls for firm size and fixed capital are

not included. Hence, in columns (3) and (4), covariates for employment

and fixed assets are added, with our variables of interest still kept

separate. Results do not vary in terms of statistical significance as

compared to the previous specifications, supporting the hypothesis that

cross-border takeovers are more inspired by market considerations. By

contrast, there is no supporting evidence for selection decisions based on

within-firm changes in productivity. A concern on the validity of these

results may arise by entering both profitability and labour productivity in

the same specification, as their effects and significance could deviate

from specifications where they are separately estimated. Therefore, we

test this by running estimations reported in columns (5) and (6), which

incorporate firm profitability and labour productivity in the same model.

Analogously to previous estimates, results remain stable suggesting that

MNEs tend to select more profitable domestic firms. As evidenced by

results in columns (1) to (4), the statistical insignificance of the

coefficient on firm labour productivity as a determinant of cross-border

takeovers in columns (5) and (6) cannot be associated with the

simultaneous inclusion of the profitability measure. Moreover, while it

could be argued that firm profitability captures an effect similar to that of

labour productivity, we have shown in the data section that their

correlation coefficient is particularly modest in magnitude. We consider

the coefficient on profitability in column (6) as our preferred baseline

estimate since this is our most extended specification. This denotes that

firms experiencing a one percentage point increase in profitability have a

probability of 0.038% of being acquired or, equivalently, a one standard

deviation increase in lagged profitability is on average associated with a

154

0.8% higher probability of being acquired by a foreign MNE in any given

year. This latter figure should be interpreted bearing in mind that only

0.15% of firms are acquired in our full sample, as shown by the numbers

reported in Table 1. Hence, the magnitude of the effect appears to be

nontrivial.

These baseline results tend to corroborate the notion that MNEs

select domestic companies that exhibit notable within-firm changes in

profitability, after potential waves of cross-border takeovers as well as

trends of corporate activity in specific industries and countries are

controlled for. This implies that domestic firms experiencing above-

average increases in their profitability are targeted by MNEs in the

following year. This provides some preliminary support to the hypothesis

that cross-border acquisitions are associated with a market entry

rationale, according to which MNEs aim at securing a solid position in

foreign locations through the acquisition of a profitable domestic

company in order to access new or larger market opportunities.

As it is mentioned above, a one year lag in the measures of

profitability and labour productivity appears reasonable according to the

evidence on the typical negotiation time required for acquisitions (Fich et

al., 2011). To a closer inspection this circumstance is also corroborated

by our data: indeed, by exploiting time information about acquisitions,

we find that 90% of transactions in our sample are rumoured or

announced in the same calendar year in which they are eventually

completed. This figure increases to 98% when also including acquisitions

that are rumoured or announced in a specific calendar year and they are

successfully completed in the following year. Therefore, in terms of

timing, cross-border takeover decisions appear to be based in most cases

on a relatively quick assessment of target firms that have recently

experienced a profitability boost. Conceptually, this common occurrence

could be considered as reasonable when cross-border acquisitions are

associated to a market access rationale: indeed, MNEs in search of new

155

or wider market opportunities plausibly tend to assess targets on their

more recent observable market performance and linkages. Also, the high

speed of the selection decision could be also underpinned by

considerations in terms of price: that is, firms with growing businesses

can become more costly in time.

4.5.2 Evidence from alternative measures of profitability and

productivity

In the previous section we have explored how firm profitability and

labour productivity affect international acquisitions by employing the

operational definitions reported in equations (1) and (2). A concern could

be that our baseline results will change with different definitions of these

measures. For instance, in considering EBIT we are incorporating

amortisation and depreciation costs in our profitability measure.

Similarly, in normalising our measures by yearly industry means we are

not accounting for the relative important role that specific firms can play

in their industry within their national boundaries. To accommodate these

and other aspects, this section offers an analysis of cross-border

takeovers by employing alternative measures of profitability and labour

productivity constructed as explained in section 3.2.

Table 5 reports the linear probability estimation results on the full

sample of domestic firms by adopting these new measures. In Panel A,

we alternate different proxies of labour productivity, while firm

profitability enters the model as specified in equation (1). Similarly, Panel

B includes labour productivity as constructed in equation (2) combined

with alternative proxy variables for firm profitability. All specifications

include covariates for firm size in terms of employment and fixed assets

as well as a full set of year dummies, industry trends, country-year

dummies and time invariant firm effects. In this respect, estimated

156

coefficients in Table 5 are directly comparable with our baseline

estimates.

[Table 5 here]

Results in Panel A support the idea that selection decisions of MNEs

are associated with the search for domestic firms with a strong ability to

make successful business. This result remains stable across

specifications when different measures of labour productivity are

employed. The latter, similarly to the baseline results, does not exhibit

statistically significant coefficients in columns (1) to (3) regardless of the

way in which the measure is constructed. As matter of fact, substituting

firm value added with turnover as well as fragmenting yearly industry

means by country constantly provides the same non-significant

estimates on labour productivity. In a similar vein, Panel B reports

estimation results that corroborate further the hypothesis that cross-

border acquisitions are influenced by the increasing success of domestic

firm boost in profitability, while MNEs do not seem to be sensitive to the

opportunity to access the productive assets of potential target

companies, ceteris paribus. The statistical relevance of firm profitability is

also robust to different operational definitions as suggested by columns

(4) to (6), where the significance level is maintained between 1% and 5%.

Interestingly, the magnitude of the effect is similar to that in our

baseline estimates, with the exception of EBITDA-based measures of

profitability, which exhibit a stronger effect. This may reasonably suggest

that depreciation and amortisation truly depress firm earnings when they

are not excluded from the definition of profitability.

A further concern with respect to these results can be related to the

inclusion of labour productivity as a proxy for domestic firms’ valuable

productive assets instead of TFP. In fact, while labour productivity tends

to capture the incidence and relevance of the workforce in transforming

157

inputs into output, it misses by definition the role played by other factors

of production. At the same time, it is possible that firm profitability

captures within-firm variation in TFP, and this would explain the

constantly significant coefficients associated with different measures of

firm profitability. In a nutshell, excluding TFP could simultaneously

explain the statistical relevance of firm profitability and the insignificance

of labour productivity. Our empirical model described by equation (3) in

Section 4 can be easily modified to accommodate the inclusion of a

measure of TFP. Therefore, two simple measures of production function-

based TFP are generated by exploiting Orbis data and following Gal

(2013). By exploiting information on firm value added, employment and

tangible fixed assets, firm TFP is estimated as the residual of both simple

OLS and fixed effects estimations at the firm level. Hence, we normalise

these two measures of TFP by yearly industry means at the 4-digits

industry level and we lag them. We eventually obtain two variables of TFP

with a similar structure to our measures of profitability and labour

productivity. The correlation coefficient between the two measures of TFP

is 0.35, suggesting that the portion of time invariant productivity is large.

This is also evident by comparing the correlation between the two

measures of TFP and labour productivity, as described in equation (2).

The coefficient stands at 0.85 in the case of OLS residual TFP, while it

decreases to 0.31 when labour productivity is compared to the fixed

effects TFP. With respect to the potential issue that profitability may

capture some TFP-type effect, this should not constitute a concern in our

data given that the correlation between OLS residual TFP and

profitability, the latter defined in equation (1), is only 0.32 and it falls to

0.08 when considering fixed effects TFP.

[Table 6 here]

158

In Table 6, the linear probability model detailed in equation (3) is

estimated by substituting labour productivity with TFP to check the

robustness of our previous results to the inclusion of such a variable. A

first observation should be made with respect to the number of firms that

enter the regression, which falls from 306,247 to 213,776, further

justifying the adoption of labour productivity in the first place to study a

larger sample and to avoid selection issues. Similarly to Table 4, columns

(1) and (2) first report the results for productivity in isolation. Neither

version of TFP yields statistical significant coefficients, in line with the

estimates of labour productivity. Columns (3) and (4) instead reflect

previous results, with a notable role played by domestic firms’ market

linkages in shaping the selection decisions of MNEs that engage in cross-

border takeovers. Therefore, the pattern illustrated in the baseline is

further supported, and concerns associated with our preferred measures

of profitability and productivity should be, at least, mitigated by the tests

performed in this section.

4.5.3 Non-linearity in within-firm probability of foreign

acquisition

While previous sections presented baseline results and their

robustness to model specification with alternative measures of

profitability and labour productivity, this part will investigate whether

the probability of foreign acquisition that each domestic firm faces in any

given year can be considered as a non-linear function of its valuable

productive assets and its capacity to run profitable businesses. The

notion that higher productive efficiency corresponds to thriving market

performance is well established (Foster et al., 2008). Therefore, this may

reasonably suggest that the probability of foreign acquisition associated

with the market access rationale underlined by our previous results

could be particularly marked in the presence of more productive

159

domestic firms. In other words, more productive domestic companies can

be those that reasonably experience a more substantial positive within-

firm expansion in profits as compared to previous years. In this respect,

firm productivity is conceptually seen as a determinant of cross-border

acquisitions that discriminates between potential targets rather than

indicating when a domestic company is acquired. This is in line with the

empirical evidence produced in Guadalupe et al. (2012), who maintain

that MNEs cherry-pick more productive firms within industries36. In our

setting, this could explain the non-significant coefficient emerging from

within-firm variation in productivity and, at the same time, the relevant

role played by thriving firm profitability. Therefore, from an empirical

point of view, we augment the probability model in equation (3) by

entering an interaction term between firm profitability and labour

productivity in order to delve into the potential interplay between these

two firm characteristics in shaping the selection decisions of MNEs.

Results of this estimation are reported in column (1) of Table 7, which

shows that the effects of firm profitability and labour productivity do not

vary as compared to previous results. Interestingly, the interaction term

yields a positive and significant coefficient, as hypothesised. Within-firm

differences in the probability of being acquired also depend upon the

level of firm productivity, conditional on being domestically-owned before

the takeover.

[Table 7 here]

To further examine this aspect, we break our sample down at median

values of firm profitability and labour productivity. In so doing, we are

able to estimate the probability of being acquired as a function of within-

36

By contrast, the study of within-firm differences in the probability of being acquired by a MNE in

Blonigen et al. (2014) suggests that the occurrence of negative shocks in firm productivity

encourages takeovers as it lowers the price.

160

firm variation in profitability using specific sub-samples of high- (low-)

productive domestic companies. Similarly, we test the differential

relevance of firm productivity in shaping foreign acquisition decisions on

sub-samples of high- (low-) profitability enterprises.

Estimates are presented in columns (2) to (5) of Table 7. The

concentration of the statistically significant effect of profitability in the

sub-sample of domestic firms exhibiting a level of labour productivity

above the median in column (2) supports the idea that more productive

domestic companies tend to experience positive within-firm variations in

their profitability that make them systematically more appealing for

takeovers than the rest of potential targets. In this sub-sample of

domestic firms, a one standard deviation increase in lagged profitability

corresponds to a 1.4% higher probability of being acquired by a foreign

MNE in any given year37, which is a larger effect than the one retrieved in

our baseline estimates. As expected, the same does not occur in column

(4) when we analyse within-firm changes in labour productivity in the

segment of high profitability companies. As mentioned, these results

corroborate the hypothesis that a notable portion of the profitability

effect tends to be concentrated among more efficient firms, as these may

plausibly be those that easily experience a reinforcement of their

businesses over time.

4.5.4 Foreign acquisitions and technology

The profitability effect emerged in previous results may also be

associated with the technological intensity of specific industries. We

employ the Eurostat aggregations of manufacturing sectors by

technological intensity based on NACE Rev.2 in order to identify

industries characterised by different levels of technology. In so doing, we

are able to group firms into high-medium technology and medium-low

37

The standard deviation of lagged profitability in this sub-sample is equal to 2.0463.

161

technology sectors. In particular, the former category encompasses

68,477 firms grouped in 92 4-digit industries while the latter contains

237,770 companies in 200 4-digit industries. The number of cross-

border takeovers is similar across the two segments of firms, with 227

acquisitions occurring in high-medium technology sectors and 231 in

medium-low technology industries. The profitability effect on the

probability of being acquired by a MNE can plausibly be associated with

sectors that are characterised by a higher technological intensity as these

industries could require higher costs of entry and investment in R&D

(Narula and Hagedoorn, 1999). Therefore, limited competition in these

sectors could be conducive of stronger increases in firm profitability over

time. The lower number of firms in high-medium technology sectors in

our data seems to point in this direction. Interestingly, firm average size

in terms of employment of firms in high-medium technology industries is

179 employees in our sample, whilst the same dimension decreases to 61

employees when considering medium-low technology sectors. Thus,

domestic firms operating in segments of the economy where the

technological content is higher are considerably larger in size. This can

produce additional barriers to entry due to strong economies of scale in

these industries, especially in presence of transport costs and non-

homogeneous goods.

In Table 8, we estimate the probability of being acquired by

differentiating between high-medium technology and medium-low-

technology sectors. Columns (1) and (2) report results for the baseline

model while columns (3) and (4) include the interaction term between

firm profitability and labour productivity. The effect of profitability tend

to be concentrated in high-medium technology sectors, as anticipated,

while cross-border acquisitions in medium-low technology industries are

not responsive to this aspect. The coefficient in column (1) suggests that

a one standard deviation increase in firm profitability corresponds to a

2.5% higher chance of switching to foreign ownership in the following

162

year, conditional of being previously domestically-owned38. The effect of

profitability holds when the interaction enters the specification. The

latter is surprisingly non-significant when examining takeovers in high-

medium technology sectors, while it becomes weakly relevant in the sub-

sample of medium-low technology industries. This may indicate that the

(weak) effect of profitability in these sectors is present only for the most

productive firms experiencing a positive variation in earnings.

[Table 8 here]

4.5.5 Completed and majority foreign acquisitions

This section is aimed at testing the robustness of our previous results

with respect to changes in the dependent variable. The measure of time

varying foreign ownership employed so far, in fact, contains different

types of cross-border takeover. The first difference relates to acquisitions

of majority and minority stakes of the target firm. In fact, different

organisational strategies by MNEs can lead to the decision to engage into

cross-border takeovers according to different degrees of control of foreign

assets. The second difference is associated with the completion of a deal

as opposed to acquisitions that are only announced or rumoured.

[Table 9 here]

Table 9 provides evidence considering these different aspects. In

columns (1) and (2), we consider a measure of majority acquisitions,

defined as transactions resulting in a total share of foreign ownership

that is equal or larger than 50%. As a result, the total number of

acquisitions decreases to 420 from the initial 458. This limited decrease

in the number of cross-border takeovers is not due to the fact that MNEs

38

The standard deviation in this subsample is 1.9983.

163

immediately acquire a large ownership share of targets. Actually,

acquisitions of more limited shares are frequent in our sample. However,

most of these are followed by further transactions in the same calendar

year by the same MNE aimed at increasing its stake of ownership. In

these cases, we consider only the last operation of acquisition, often

resulting in a majority takeover. By contrast, when more operations

between the same acquirer and target span over different calendar years,

we consider the first operation only. Columns (3) and (4), instead, report

estimation results for completed operations. As mentioned, some

acquisitions are only announced or rumoured, whereas completed

acquisitions amount to 416 transactions. Finally, in columns (5) and (6),

we simultaneously combine information on majority and completed

operations, thus obtaining 387 cross-border acquisitions. Results

continue to support the notion that takeovers are associated with market

access considerations via domestic firms experiencing thriving business

conditions. Furthermore, similarly to previous results, columns (2), (4)

and (6) report that this effect is also mediated by firm productivity: that

is, when a domestic company is more efficient, within-firm expansion in

profitability tends to be associated with a higher chance of being

acquired in a given year. The magnitude of the estimated coefficients is

also in line with previous results.

4.5.6 Evidence from acquisition targets only

In the previous sections, we employed a full sample containing both

firms that are acquired at some point over the period 1997-2013 and

firms that remain domestically-owned over the whole time span. By

contrast, in this part the sample is restricted to domestic firms that are

acquired by foreign MNEs in a certain year, similarly to the empirical

strategy of Blonigen et al. (2014). Therefore, we test whether within-firm

variation in profitability and productivity also explain differences in the

164

probability of foreign acquisition in the group of targeted domestic

companies. This empirical approach can also be informative of the timing

of foreign takeovers, considering that all firms in the sample are acquired

by the end of the sample period.

[Table 10 here]

Table 10 presents the estimation results based on data on 268 firms

acquired over the period. We lose some of the acquisitions as compared

to the full sample due the generation of new variables for firm

profitability, productivity and other characteristics as well as yearly

industry means. Results still support the idea that MNEs that engage in

cross-border takeovers select domestic firms experiencing a boost in their

business performance in the form of higher profitability, while firm

productivity does not play a relevant role. The significance level of the

coefficients on firm profitability, however, ranges between 5% and 10%.

Furthermore, we do not detect any significant interaction effect between

firm profitability and productivity. These differences are probably due to

a more limited within-firm variation in profitability and productivity in

the sample restricted to acquisition targets only, as compared to the full

sample. Overall, however, results in Table 10 still support the baseline

estimates as well as the hypothesis in favour of market access

considerations as the fundamental element that informs the selection

decisions of MNEs engaging in acquisitions across borders.

4.6 Conclusions

The relevance of M&A over other forms of FDI has notably grown in

the last decades. This is particularly the case of FDI in advanced

economies, where the acquisition of pre-existing domestic firms is the

preferential strategy of entry of MNEs. In spite of this, academic research

165

trying to understand the selection decisions of MNEs that engage in

cross-border takeovers has lagged behind, in part as a result of lack of

information on changes in the ownership structure of companies.

Therefore, shedding light on the systematic patterns of selection that

characterise the choices of international acquirers has become

particularly urgent in both academic and policy terms. In fact, with few

recent exceptions, existing econometric studies only focus on industry-

wide or country-wide determinants of acquisitions and the micro-level

drivers of this important form of FDI remain underscored. This lack of

quantitative empirical evidence on a central feature of current

globalisation (i.e. cross-border acquisitions) represents an important

motivation developing the present chapter.

In this paper we have hypothesised that while the productivity

mechanism suggested by the literature can be a relevant driver of

acquisitions, market access considerations could be analogously

important in shaping the behaviour of MNEs. In fact, corporate strategies

can be also aimed at securing a position of strength in a foreign market

via the acquisition of a domestic firm experiencing thriving business

performance. By employing data on European firms in EU ‘old’ member

countries, we found strong evidence in favour of this second hypothesis,

while productivity motives for acquisition do not find any support in our

sample. This finding appears especially meaningful considering that EU-

15 countries are notoriously associated with inflows of FDI aimed at

accessing the large European market (Head and Mayer, 2004). Our

results are robust to different measurements of productivity and firm

profitability. Furthermore, our findings suggest that the effect of positive

within-firm variation in business conditions tends to be concentrated

among more productive firms, providing some support for the notion that

MNEs acquire more efficient firms that are capable to increase the

profitability of their business operations. As expected, the relevance of

the time varying capacity of firm to make profits is concentrated in

166

industries characterised by higher technological intensity. This is

possibly due to the higher barriers to entry and the presence of scale

economies in these sectors. Our findings are also robust to different

definitions of foreign ownership, including the acquisition of majority

stakes and the inclusion of completed transactions only. Finally, our

main results also hold when reducing the sample to include only those

firms that switch from domestic to foreign ownership during the sample

period. It is important to notice, however, that within-industry

differences in MNE performance can be associated with different

propensity and ability to accumulate knowledge, invest in R&D as well as

managerial capability (Castellani and Zanfei, 2006; Castellani and

Giovannetti, 2010). Therefore, selection in the acquisition strategies of

MNEs could be related to some extent to MNE diversity in these

underlying characteristics. While the present study is limited in this

respect, this can be considered a valuable line for future research on the

selection patterns of cross-border takeovers. Moreover, data limitation

does not allow measuring strategic assets of target firms in a neat

manner, but several measures for productivity are employed. A further

limitation is associated with the existence of explanations of international

acquisitions not directly tested in this chapter. For instance, MNEs from

emerging countries increasingly adopt knowledge augmenting strategies

by acquiring companies in developed countries (Luo and Tung, 2007).

Our data does not allow identifying a sufficient number of transactions

undertaken by this type of MNE and therefore this analysis cannot be

adequately developed from an econometric standpoint.

In a policy perspective, this paper’s findings can be considered to

delineate measures to support industrial restructuring in the EU as a

strategy of firms to maintain or increase their competitiveness. However,

policy makers should also be concerned with the risks associated to large

waves of M&A in terms of a reduction of market competition through the

167

acquisition activity of MNE, thus taking into account a reinforcement of

antitrust policies.

168

Table 4.1: Firms and acquisitions by country, 1997-2013

(1) (2) (3) (4) (5)

Country Observations Firms % Acquisitions %

A. Full sample

Austria 3,345 1,300 0.42 6 1.31

Belgium 15,245 3,254 1.06 35 7.64

Denmark 253 42 0.01 0 0.00

France 129,674 38,050 12.42 63 13.76

Finland 23,273 6,672 2.18 17 3.71

Germany 68,970 23,013 7.51 79 17.25

Greece 18 8 0.00 0 0.00

Italy 325,555 91,964 30.03 93 20.31

Ireland 1,407 490 0.16 2 0.44

Netherlands 1,063 320 0.10 2 0.44

Portugal 79,640 24,556 8.02 6 1.31

Spain 419,717 90,231 29.46 49 10.70

Sweden 70,720 15,958 5.21 28 6.11

United Kingdom 39,015 10,389 3.39 78 17.03

Total 1,177,895 306,247 100.00 458 100.00

B. Restricted sample

Austria 11 3 1.12 3 1.12

Belgium 48 22 8.21 22 8.21

Denmark 0 0 0.00 0 0.00

France 69 25 9.33 25 9.33

Finland 25 10 3.73 10 3.73

Germany 109 44 16.42 44 16.42

Greece 0 0 0.00 0 0.00

Italy 171 59 22.01 59 22.01

Ireland 0 0 0.00 0 0.00

Netherlands 6 2 0.75 2 0.75

Portugal 1 1 0.37 1 0.37

Spain 107 34 12.69 34 12.69

Sweden 68 19 7.09 19 7.09

United Kingdom 144 49 18.28 49 18.28

Total 759 268 100.00 268 100.00

Notes: A) Columns 1, 2, and 3 are based on Orbis data, while columns 4 and 5 are based on

Zephyr.

169

Table 4.2: Correlation between measures of profitability and labour productivity

PR1 PR2 PR3 PR4 LP1 LP2 LP3 LP4 EM AS

Full sample

ln (ebit/assets) t-1 (PR1) 1

ln (ebitda/assets) t-1 (PR2) 0.89 1

ln (ebit/assets) by country t-1 (PR3) 0.72 0.66 1

ln (ebitda/assets) by country t-1 (PR4) 0.66 0.72 0.90 1

ln (value added/empl.) t-1 (LP1) 0.13 0.12 0.08 0.07 1

ln (turnover/empl.) t-1 (LP2) 0.13 0.12 0.08 0.07 0.73 1

ln (value added/empl.) by country t-1 (LP3) 0.10 0.09 0.12 0.11 0.83 0.58 1

ln (turnover/empl.) by country t-1 (LP4) 0.11 0.10 0.13 0.11 0.62 0.82 0.70 1

ln employmentt-1 (EM) -0.04 -0.04 -0.11 -0.11 0.15 0.13 0.04 0.01 1

ln assetst-1 (AS) -0.31 -0.34 -0.37 -0.39 0.38 0.34 0.28 0.22 0.75 1

Restricted sample

ln (value added/empl.) t-1 (LP1) 0.29

Notes: PR stands for profitability and LP stands for labour productivity. All variables are normalised by industry

mean as explained in the relative section.

170

Table 4.3: Descriptive statistics

Obs Mean

Std.

Dev. Obs Mean

Std.

Dev.

Full sample Restricted sample

ln (ebit/assets) t-1 1177895 -1.899 2.078 759 -0.2647 1.1888

ln (ebitda/assets) t-1 1124238 -1.372 1.638

ln (ebit/assets) by country t-1 1002614 -1.327 1.989

ln (ebitda/assets) by country t-1 1037000 -0.953 1.614

ln (value added/empl.) t-1 1177895 -0.161 0.663 759 -0.0273 0.4052

ln (turnover/empl.) t-1 1149393 -0.430 0.898

ln (value added/empl.) by country t-1 1177084 -0.100 0.592

ln (turnover/empl.) by country t-1 1149392 -0.256 0.767

ln employment t-1 1177895 -1.109 1.482 759 -0.5768 1.3810

ln assets t-1 1177895 -1.900 2.231 759 -0.9005 1.845

Notes: PR stands for profitability and LP stands for labour productivity. All variables are

normalised by industry mean as explained in section 3.2.

171

Table 4.4: Probability of foreign acquisition

(1) (2) (3) (4) (5) (6) Dep Var: Foreign

ownership

ln Profitability t-1 0.0035***

0.0040***

0.0037** 0.0038**

(0.0013)

(0.0015)

(0.0013) (0.0015)

ln Labour productivity t-1

0.0063

0.0075 0.0048 0.0045

(0.0073)

(0.0084) (0.0073) (0.0084)

ln Employment t-1

-0.0011 0.0030

0.0008

(0.0056) (0.0067)

(0.0067)

ln Assets t-1

0.0033 -0.0007

0.0026

(0.0032) (0.0028)

(0.0032)

Observations 1,177,895 1,177,895 1,177,895 1,177,895 1,177,895 1,177,895

Clusters 306,247 306,247 306,247 306,247 306,247 306,247

R-squared 0.46 0.46 0.46 0.46 0.46 0.46

adj. R-squared 0.26 0.26 0.26 0.26 0.26 0.26

Year FEs Y Y Y Y Y Y

Country-year dummies Y Y Y Y Y Y

Industry trends Y Y Y Y Y Y

Firm FEs Y Y Y Y Y Y

Notes: A) Firm-level clustered standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1. B) All

variables are normalised by industry means computed yearly at NACE 4-digits level.

172

Table 4.5: Alternative measures for profitability and labour productivity

(1) (2) (3) (4) (5) (6)

Dep Var: Foreign ownership A. Alternative measures of labour productivity

Profitability=ln(ebit/assets) t-1 0.0036** 0.0037** 0.0034**

(0.0015) (0.0015) (0.0015)

Labour productivity = ln (turnover/empl.) t-1 0.0037

(0.0066)

ln (value added/empl.)

0.0057 by country t-1

(0.0085)

ln (turnover/empl.)

0.0083 by country t-1

(0.0089)

B. Alternative measures of profitability

Labour productivity = ln (value added/employment) t-1

0.0035 0.0142 0.0136

(0.0091) (0.0087) (0.0085)

Profitability = ln (ebitda/assets) t-1

0.0062***

(0.0022)

ln (ebit/assets) by country t-1

0.0040**

(0.0017)

ln (ebitda/assets) by country t-1

0.0051**

(0.0020)

C. Both panels

ln Employment t-1 0.0009 0.0012 0.0030 -0.0006 0.0078 0.0076

(0.0066) (0.0066) (0.0074) (0.0070) (0.0077) (0.0074)

ln Assets t-1 0.0028 0.0025 0.0022 0.0041 0.0027 0.0031

(0.0031) (0.0032) (0.0032) (0.0035) (0.0036) (0.0037)

Observations 1,149,393 1,177,084 1,149,392 1,124,238 1,002,614 1,037,000

Clusters 300,389 306,193 300,389 300,174 290,959 293,043

R-squared 0.46 0.46 0.46 0.47 0.50 0.49

adj. R-squared 0.27 0.26 0.27 0.27 0.30 0.29

Year FEs Y Y Y Y Y Y

Country-year dummies Y Y Y Y Y Y

Industry trends Y Y Y Y Y Y

Firm FEs Y Y Y Y Y Y

Notes: A) Firm-level clustered standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1. B) All variables are normalised by industry means computed yearly at NACE 4-digits level. Where specified, industry means are also calculated by country.

173

Table 4.6: Foreign acquisitions and total factor productivity

(1) (2) (3) (4) Dep Var: Foreign

ownership

ln Profitability t-1

0.0061*** 0.0059**

(0.0023) (0.0023)

ln TFP ols t-1 0.0172

0.0112

(0.0141)

(0.0140)

ln TFP fe t-1

0.0237

0.0165

(0.0152)

(0.0150)

ln Assets t-1 0.0040 0.0030 0.0088 0.0079

(0.0056) (0.0054) (0.0060) (0.0058)

ln Employment t-1 0.0045 -0.0008 0.0007 -0.0026

(0.0131) (0.0110) (0.0130) (0.0109)

Observations 662,910 662,910 662,910 662,910

Clusters 213,776 213,776 213,776 213,776

R-squared 0.51 0.51 0.51 0.51

adj. R-squared 0.28 0.28 0.28 0.28

Year FEs Y Y Y Y

Country-year dummies Y Y Y Y

Industry trends Y Y Y Y

Firm FEs Y Y Y Y

Notes: A) Firm-level clustered standard errors in parentheses. ***

p<0.01, ** p<0.05, * p<0.1. B) All variables are normalised by industry

means computed yearly at NACE 4-digits level.

174

Table 4.7: Interaction effect between firm profitability and labour productivity

(1) (2) (3) (4) (5)

Full

sample

High

productivity

Low

productivity

High

profitability

Low

profitability Dep Var: Foreign

ownership

(>50%) (<50%) (>50%) (<50%)

ln Profitability t-1 0.0046*** 0.0070** 0.0024

(0.0015) (0.0033) (0.0027)

ln Labour

productivity t-1 0.0105

0.0262 -8.00e-05

(0.0075)

(0.0184) (0.0132)

Interaction t-1 0.0038**

(0.0017)

ln Employment t-1 0.0003 4.75e-05 -0.0178 0.0113 -0.0069

(0.0057) (0.0119) (0.0148) (0.0115) (0.0127)

ln Assets t-1 0.0027 0.0081 -0.0008 -0.0013 0.0012

(0.0027) (0.0062) (0.0062) (0.0054) (0.0077)

Observations 1,177,895 502,265 503,006 502,252 503,033

Clusters 306,247 245,122 245,386 245,123 245,391

R-squared 0.46 0.67 0.71 0.65 0.72

adj. R-squared 0.26 0.35 0.44 0.31 0.45

Year FEs Y Y Y Y Y

Country-year dummies Y Y Y Y Y

Industry trends Y Y Y Y Y

Firm FEs Y Y Y Y Y

Notes: A) Firm-level clustered standard errors in parentheses. *** p<0.01, ** p<0.05, *

p<0.1. B) All variables are normalised by industry means computed yearly at NACE 4-digits

level.

175

Table 4.8: Probability of foreign acquisition by technological class

(1) (2) (3) (4) Dep Var: Foreign

ownership High-Medium

tech. Medium-Low

tech. High-Medium

tech. Medium-Low

tech.

ln Profitability t-1 0.0124** 0.0013 0.0141** 0.0019

(0.0053) (0.0013) (0.0061) (0.0015)

ln Labour productivity t-

1 0.0043 0.0051 0.0144 0.0101

(0.0224) (0.00741) (0.0244) (0.0073)

Interaction t-1

0.0064 0.0032*

(0.00610) (0.00174)

ln Employment t-1 0.0057 0.0009 0.0053 0.0003

(0.0185) (0.0057) (0.0185) (0.0057)

ln Assets t-1 0.0107 0.0003 0.0111 0.0004

(0.0093) (0.0031) (0.0094) (0.0031)

Observations 272,394 905,501 272,394 905,501

Clusters 68,477 237,770 68,477 237,770

R-squared 0.43 0.48 0.43 0.48

adj. R-squared 0.23 0.30 0.23 0.30

Year FEs Y Y Y Y

Country-year dummies Y Y Y Y

Industry trends Y Y Y Y

Firm FEs Y Y Y Y

Notes: A) Firm-level clustered standard errors in parentheses. *** p<0.01, ** p<0.05, *

p<0.1. B) All variables are normalised by industry means computed yearly at NACE 4-digits

level.

176

Table 4.9: Completed and majority acquisitions

(1) (2) (3) (4) (5) (6)

Dep Var: Foreign

ownership Majority Completed Completed majority

ln Profitability t-1 0.0035** 0.0044*** 0.0033** 0.0041** 0.0032** 0.0041***

(0.0014) (0.0017) (0.0014) (0.0016) (0.0014) (0.0016)

ln Labour

productivity t-1 0.0023 0.0092 0.0043 0.0104 0.0027 0.0092

(0.0082) (0.0085) (0.0082) (0.0084) (0.0080) (0.0083)

Interaction t-1

0.0043**

0.0038**

0.0041**

(0.0019)

(0.0018)

(0.0018)

ln Employment t-1 0.0005 0.0001 0.0023 0.0018 0.0016 0.0010

(0.0065) (0.0065) (0.0065) (0.0065) (0.0064) (0.0064)

ln Assets t-1 0.0019 0.0020 0.0017 0.0019 0.0013 0.0014

(0.0031) (0.0031) (0.0031) (0.0031) (0.0030) (0.0030)

Observations 1,177,895 1,177,895 1,177,895 1,177,895 1,177,895 1,177,895

Clusters 306,247 306,247 306,247 306,247 306,247 306,247

R-squared 0.46 0.46 0.46 0.46 0.46 0.46

adj. R-squared 0.27 0.27 0.27 0.27 0.27 0.27

Year FEs Y Y Y Y Y Y

Country-year

dummies Y Y Y Y Y Y

Industry trends Y Y Y Y Y Y

Firm FEs Y Y Y Y Y Y

Notes: A) Firm-level clustered standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.

B) All variables are normalised by industry means computed yearly at NACE 4-digits level.

177

Table 4.10: Restricted sample

(1) (2) (3)

Dep Var: Foreign ownership

ln Profitability t-1 4.604** 3.994* 4.015*

(2.253) (2.312) (2.297)

ln Labour productivity t-1 -3.180 -0.944 -0.838

(7.433) (7.833) (8.413)

Interaction t-1

0.182

(3.539)

ln Employment t-1 2.658 2.637

(4.440) (4.429)

ln Assets t-1 -3.798 -3.792

(3.675) (3.673)

Observations 759 759 759

Clusters 268 268 268

R-squared 0.63 0.63 0.63

adj. R-squared 0.57 0.57 0.57

Year FEs Y Y Y

Country-year dummies Y Y Y

Industry trends Y Y Y

Firm FEs Y Y Y

Notes: A) Firm-level clustered standard errors in

parentheses. *** p<0.01, ** p<0.05, * p<0.1. B) All

variables are normalised by industry means computed

yearly at NACE 4-digits level.

178

Part III: The Impact of FDI on

Recipient Economies

179

Chapter 5 – Inward FDI and Local

Innovative Performance. An empirical

investigation on Italian provinces

5.1 Introduction

In the current wave of globalisation of the world economy it is widely

acknowledged that foreign direct investment (FDI) plays a growing and

primary role (WTO, 1996; Dicken, 2007). UNCTAD (2012) shows that the

volume of FDI has dramatically risen in the last twenty years, with an

increase in world FDI inward stock of about 2 millions of dollars to more

than 20 millions.

Not surprisingly, policy makers in most countries place great

emphasis on the potential benefits that may stem from the attraction of

FDI. The view that attracting foreign subsidiaries of multinational

enterprises (MNEs) will yield great advantages to recipient economies is

grounded in the belief that some positive knowledge externalities arise

from foreign activities and spread to domestic firms. Beside of several

potential benefits, the increase of domestic productivity and the transfer

of more advanced technology are frequently considered as the main

rationale for integrating measures of attraction of FDI in local economic

development policies. In this respect, the idea that knowledge plays a

fundamental role in the process of growth is deeply rooted in economic

theory, which assigns a crucial role to innovation and its diffusion in the

economic performance of nations (Grossman and Helpman, 1991).

Nevertheless, it is not entirely clear whether FDI concretely benefits

recipient economies. Despite the large amount of studies in this field and

its relevance for public policies, evidence on FDI-induced knowledge

180

externalities remains inconclusive and empirical exercises frequently

offer mixed suggestions (Smeets, 2008).

By employing Italian manufacturing data to answer the question

whether inward FDI benefit the innovative performance of recipient

economies, this paper will attempt to add some new evidence to the

literature on the impact of FDI. There are a number of elements that

make this empirical exercise different from the bulk of previous research.

Firstly, the impact of knowledge externalities associated to FDI is

investigated on direct measure of innovation, namely, patent data. To the

best of our knowledge, few papers adopt such an indicator (Cheung and

Lin, 2004) while the literature is dominated by studies based on broader

measures of economic performance such as total factor productivity (TFP)

of domestic firms, labour productivity or growth rate. Secondly, FDI are

also measured with a direct indicator. Indeed, while most studies use

several proxies for the presence of foreign firms into the host economy,

this paper employs the real inflow of foreign capital in Italy. This provides

a more detailed measure of the actual magnitude of the activities carried

out by foreign enterprises. Thirdly, FDI-induced knowledge externalities

are underexplored in the case of Italy, with few notable exceptions

represented by recent contributions (Castellani and Zanfei, 2003; 2007;

Benfratello and Sembenelli, 2006). The Italian case is instead very

interesting for the well-known geographical dualism of the Italian

economy. Finally, the occurrence of knowledge spillovers is investigated

along provincial lines (NUTS-3), that is, at a geographical scale that is

rarely adopted in the literature mainly due to lack of data. This allows

estimating a more precise effect by reducing the potential ecological

fallacy39 and also taking into appropriate consideration the existence of

39

In its simplest definition the ecological fallacy may be interpreted as error of deduction that involves

deriving conclusions about a certain observation solely on the basis of an analysis of broader group data. In

the case of this analysis the inference on the impact of FDI on local innovative performance may be

181

spatial disparities in both inward FDI and innovation performance.

Results reveal that local production systems do benefit from knowledge

externalities generated by FDI in Italy. Our finding also passes a fair

number of checks suggesting that local innovative performance relies on

both internal and external sources of knowledge.

The paper is organized as follows: Section 2 reviews the existent

literature devoted to the economic rationale of the impact of FDI on

innovation. Section 3 describes data while Section 4 introduces the main

methodological challenges associated to the estimation of the causal

effect of FDI on innovation and presents in detail the identification

strategy adopted. Section 5 discusses the main findings while in Section

6 the robustness of results is checked. Finally, concluding remarks and

policy considerations are developed in Section 7.

5.2 Conceptual background and literature

review

Traditionally, the literature on FDI spillovers implicitly assumes that

MNEs have more advanced technology than most domestic firms. Hence,

the entry of foreign affiliates into an economy is believed to benefit local

firms by providing them with a number of advantages not available

domestically, ranging from new technologies to market opportunities.

The “superiority” of foreign firms has been firstly theorised within the

industrial organisation literature by Hymer (1976/1960)40. Domestic

firms have general advantages linked to better information about the

inaccurate if performed at a broader geographical level of analysis for two key reasons due to the extreme

heterogeneity in terms of structure, composition and absorptive capacities of different local areas

(Gagliardi, 2015). 40

Hymer’s seminal theory is contained in his 1960 doctoral dissertation which was published posthumously

in 1976.

182

national market, the language and the legal and political system. Thus,

firms wishing to operate in foreign markets need to overcome domestic

competition by increasing their efficiency through the acquisition of firm-

specific advantages. These include the capacity to access factors of

production at lower cost, product differentiation and the availability of

more advanced knowledge. This initial conceptualisation is further

supported by Dunning (1980), who theorises the existence of ownership-

specific advantages possessed by some firms that decide to internalise

them and to locate in foreign markets as a way to maximize their

productive efficiency in a world of imperfect competition and uncertainty.

This literature suggests that FDI occurs when firms possess own assets

and find more profitable to internalise the use of such advantages rather

than selling or sub-contracting them to other firms. At the same time,

these firms decide to locate in foreign countries where specific location

factors allow for a better exploitation of their ownership advantages.

More recently, but in a similar vein, scholars suggest that MNEs are

more productive and innovative than domestically-oriented firms

(Criscuolo et al., 2010). Indeed, it is widely acknowledged that MNEs

tend to invest large amounts in R&D, generating a notable share of global

knowledge (Castellani and Zanfei, 2006; Dicken, 2007; McCann and Acs,

2009).

Given the alleged superiority of technology and assets of MNEs, it is

commonly believed that when a foreign subsidiary locates in a new

market some knowledge spills over to domestic firms. The idea that FDI

may benefit host economies through spillover effects is empirically

explored since the 1970s. Early works find a positive relationship

between the foreign presence in a host economy and the performance of

domestic firms (Caves, 1974, Globerman, 1979, Blomström and Persson,

1983).

Since the 1990s empirical works have increasingly refined along with

improvements in the quality of data. In general, recent works try to open

183

what Görg and Strobl (2005) call “the black box” of spillover effects from

FDI. In other words, researchers have started to explore both

theoretically and empirically a number of specific mechanisms through

which the presence of foreign activities may benefit domestic firms

(Blömstrom and Kokko, 1998; Liu et al., 2000; Liu et al., 2001; Saggi,

2002; Harris, 2009). Research indicates that the nature of these

channels of knowledge transmission is essentially dual for interactions

between domestic and foreign firms occur at both intra- and inter-

industry level. Intra-industry (or horizontal) interactions between foreign

and domestic firms may lead to knowledge leakages through a variety of

mechanisms. Some scholars suggest that demonstration effects play a

great role in knowledge transmission whenever domestic firms are

exposed to the superior technology of MNEs subsidiaries (Castellani and

Zanfei, 2003; Görg and Greenaway, 2004; Crespo and Fontoura, 2007,

Smeets, 2008; Monastiriotis and Alegria, 2011). Part of the literature

argues that intra-industry spillovers may be denser in more competitive

markets. The competitive pressure caused by the entry of foreign firms

may act as an incentive for domestic firms to use available resources and

existing technology more efficiently (Blomström, 1989; Wang and

Blomström, 1992) as well as speeding up the process of adoption of new

technologies (Görg and Greenaway, 2004). Finally, intra-industry

spillovers have been analysed looking at labour mobility (Fosfuri, Motta

and Rønde, 2001) as well as pre-existing regional innovativeness (Huang

et al., 2012).

Inter-industry (or vertical) interactions between foreign and domestic

firms appear to be more witting than intra-industry dynamics. As a

matter of fact, when firms operate in different industrial segments that

are vertically connected with each other, they can intentionally establish

backward and forward linkages. From an empirical point of view a

number of evidences have been provided in support of the existence of

184

valuable inter-industry spillovers working through backward and forward

linkages (Blalock, 2001, Ernst and Kim, 2002, Crespo and Fontoura,

2007, Javorcik, 2004, Javorcik and Spatareanu, 2008, 2009, Bitzer et

al., 2008, Blalock and Gertler, 2008, Markusen and Venables, 1999,

Castellani and Zanfei, 2006, Crespo and Fontoura, 2007).

Beside this large body of literature, it is worth mentioning that some

recent contributions highlight that the origin of foreign investment is a

relevant aspect for a full understanding of FDI-induced spillovers. In fact,

while it is customary to conceive MNEs as endowed with superior

knowledge as compared to domestic firms, this is something that is

intimately connected with the evidence associated to MNEs from

industrial countries. Nevertheless, in recent years, the growing

importance of emerging countries in the global arena (e.g. BRICS), is

accompanied by a spur in FDI originating from developing countries (Luo

and Tung, 2007). Specifically, MNEs from this group might not be

endowed with superior technological attributes and, instead, their

internationalisation strategies in foreign locations (i.e. industrial

countries) are likely to be oriented towards knowledge augmentation and

the acquisition of strategic resources (Chen and Chen, 1998; Mathews,

2002; Luo and Tung, 2007). In this respect, there is some evidence that

when domestic firms are acquired by MNEs from developing countries,

the former suffer large decreases in employment, sales and labour

productivity as compared to takeovers undertaken by MNEs from

industrial countries (Chen, 2011).

5.3 Data

185

Data used for the analysis is collected from different and

complementary sources aggregated at provincial level41. Due to the

nature of the data and, particularly, to the characteristics of our

dependent variable (i.e. patent data) and main regressor of interest (i.e.

FDI inflows), the analysis will be restricted to the manufacturing sector.

All variables are taken in logarithms and averaged across the period

2001-200642.

Innovative performance - The dependent variable is defined as the

provincial share of patents on provincial GDP and it is provided by the

OECD REGPAT database containing detailed patent data at NUTS-3

level. Despite some well-known limitations associated with the non-

patentability of some inventions, the difficulties in accounting for the

differentiated degree of novelty of patented products (not all patented

products are equally ‘new’ and/or valuable) and their potential sectoral

bias, patent data remains a reliable measure of innovative output since it

provides comparable information on inventions across different regions

and a broad range of technological sectors (OECD, 2001; Sedgley and

Elmslie, 2004). Moreover, it is worth noting that we consider patents

filled by applicants rather than inventors since MNEs tends to apply for a 41

Note that we consider 103 provinces over the total number of 107 because of the lack of data on the 4

more recently-created Sardinian provinces of Olbia-Tempio, Medio Campidano, Ogliastra and Carbonia

Iglesias. Note also that provinces are administrative areas in Italy rather than functional units. Alternative

geographies include “Sistemi Locali del Lavoro” that are functional labour markets areas defined based

commuting flows. However data for these units are more limited and available for much shorter time series.

In addition to that it is worth noting that the majority of existing studies in Italy adopts either NUTS2 or

NUTS3 areas as spatial unit of analysis. This facilitates the comparability of results. 42

Patent data at the NUTS3 level are in principle available for a longer time series; however data on control

variables at the provincial level prior to 2001 are unavailable. Even though still relatively limited, the

coverage of a six year period is a significant improvement on the existent quantitative literature on the

determinants of innovation in the Italian provinces. All existing studies cover a similar or shorter time span.

For example Cainelli et al. (2005) looking at the role of social and institutional factors on the innovative

performance of Industrial districts in Emilia Romagna cover the 2002-2007 period; in a similar vein

Laursen and Masciarelli (2007), whose analysis is focused on larger geographical units (NUTS-2 Regions),

still cover a shorter time interval (2001-2003). More specifically related to the impact of FDI on

productivity in the Italian case, Castellani and Zanfei (2003) use firm level data for the period 1993-1997

while Castellani and Zanfei (2007) uses a time span 1992-1997.

186

patent from their headquarters, even when a patent is invented in a

different region (Verspagen and Schoenmakers, 2000). This measure, in

line with the existing literature, is likely to limit any concern related with

the noise associated to patent applications filled by inventors who are

resident in the recipient province and employed in foreign subsidiaries.

In other words, our measure of innovation does not include the patenting

activity of MNEs, which would bias our estimation of knowledge

externalities.

FDI Inflows - Data on inward FDI comes from the Balance of

Payments of the Bank of Italy. The database provides detailed data on

financial flows by province and sector. This represents a key advantage

over the existing literature using indirect proxies for the presence of

MNEs (e.g. share of foreign employment, share of foreign enterprises)

instead of direct measures of flow. Figure 5.1 shows the FDI trend for the

period 2001-2006. The upper left graph plots the national share of FDI

inflows showing an increasing amount of foreign capital into the Italian

economy over the whole period. However, when trends by macroregion

are taken into account it is evident that the national aggregate is driven

by Northern regions while the contribution of the South remains

negligible. This preliminary evidence suggests the existence of a relevant

and significant self-selection of FDI into more productive areas making

more urgent the need of addressing reverse causality.

[Figure 5.1 here]

Innovative Inputs - Controls for the amount of private investments in

R&D and the share of graduates in science and technology on total

population are provided by ISTAT and are available at regional level

(NUTS-2).

187

Additional Regressors - Further controls include the share of

employment in manufacturing in each province as proxy for

specialization, the share of long term unemployment as proxy for the

characteristics of the local labour market and population density as

proxy for agglomeration economies. All these additional regressors are

provided by ISTAT at NUTS-3 level. Furthermore, a full set of macro-

regional dummies defined at NUTS-1 level is included to control for

unobserved regional characteristics.

The detailed description of variables used in the analysis is reported

in Table 5.1.

[Table 5.1 here]

5.4 Methodology

The estimation of the relationship between FDI and innovation

implies a number of methodological issues. First of all, it has to be

considered that the impact of FDI on local innovation is unlikely to be

recoverable on a yearly basis. The existence of a certain time lag between

the localization of a new business activity and the emergence of a related

innovative outcome is perfectly reasonable, both if the impact of FDI

passes through the innovative activities performed by the new firm and if

this impact is instead mediated by an externality mechanism. This

concern is exacerbated by the nature of our innovation variable. Despite

adopting patent applications43 as key measure for innovative activities,

the granting procedure may require a certain amount of time before

being formalized.

Moreover, the possibility to exploit the panel dimension is prevented

by an additional consideration. Unfortunately, some of our relevant

controls, in particular the amount of investments in R&D and the share

43

Defined as the OECD as the closest data to the inventive process

188

of graduates, are only available at regional level (NUTS-2). This implies

that a certain degree of measurement error is likely to affect our

estimation and to lower the credibility of our results. Finally, due to the

limited time dimension of the panel, ranging from 2001 to 2006, the

within variation in our sample may be insufficient to identify the effect of

our regressor of interest (Baltagi, 2005).

Finally, it is also worth emphasizing that externalities are particularly

difficult to identify since externalities, by their very nature, “leave no

obvious paper trail by which they can be tracked or measured”

(Duranton, 2006, p.26). Nonetheless spatially aggregated measures of

FDI should provide a better proxy of the total effect over and above its

direct component (see Moretti, 2004) that in the case of this paper may

be associated to the innovative contribution of the individual foreign

subsidiary.

Taking into account all these aspects, the analysis of the impact of

FDI on local innovation is developed by adopting a between-groups

approach based on ordinary least squares (OLS). This implies using time

averages of data for the time interval 2001-2006 (group means)44.

The estimated equation is defined as a place based Knowledge

Production Function (KPF) at provincial level (Crescenzi et al., 2013),

where inward FDI is included as an additional regressor and externalities

associated to FDI are modelled according to a spatial correlation

approach.

The equation of interest will therefore take the following form:

𝐼𝑛𝑛𝑜𝑣𝑎𝑡𝑖𝑣𝑒 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒𝑖𝑡 = 𝛽0 + 𝛽1𝐾𝑖𝑡 + 𝛽2𝐿𝑖𝑡 + 𝛽3𝐹𝐷𝐼 𝐼𝑛𝑓𝑙𝑜𝑤𝑠𝑖𝑡 + 𝛽𝑋𝑖𝑡 + 휀𝑖𝑡 (1)

44

As acknowledged by the existing literature the between-groups estimator is more suitable to address

issues related to measurement error as compared to standard panel techniques such as random or fixed

effects estimators.

189

where 𝐼𝑛𝑛𝑜𝑣𝑎𝑡𝑖𝑣𝑒 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒𝑖𝑡 is the share of patents by applicant over

provincial GDP in province i at time t, 𝐾𝑖𝑡is the share of private

investments in R&D, 𝐿𝑖𝑡 is the share of graduates in science and

technology, 𝐹𝐷𝐼𝑖𝑡 is our regressor of interest, namely FDI inflow as share

of provincial GDP, 𝑋 is a vector of provincial controls including the share

of employment in manufacturing as proxy for specialization, long term

unemployment share, population density and a full set of macro-regional

dummies.

Another traditional methodological issue that has been highlighted in

the existing literature is the potential reverse causality between FDI and

innovation. Our key hypothesis is that FDI affect local innovative

performance contributing to enrich the local knowledge-base and

generating positive spillovers through virtuous cycles of cooperation and

competition. However, the sign of the relation may indeed be reverse: FDI

may be more attracted by areas showing successful innovative

performance since, as profit-maximizing agents, firms may have an

economic incentive to locate in successful areas and to exploit the

advantages associated with local favourable conditions. This is a

particularly relevant concern in the case of MNEs aiming to tap into local

capabilities and to benefit from local competitive advantages, which

would imply the risk of overestimating the impact of FDI. On the other

side, the effectiveness of new financial investments as carriers of novel

information and best practices may be negatively affected by a local

environment that is not able to absorb and transform these inputs into

innovation. This further entails that in the case of deprived areas or

locations characterized by relevant deficits in terms of local absorptive

capacities we may underestimate the impact of FDI. As emphasized by

190

previous research in a multilevel analysis the sign of the potential bias is

not straightforward (Haskel et al., 2007).

Most recent papers attempt to disentangle the true effect of FDI either

exploiting GMM techniques (Benfratello and Sembenelli, 2006; Driffield,

2006; Crespo et al., 2009) or through an IV approach (Haskel et al.,

2007; Crescenzi et al., 2013).

To recover predictions about the genuine causality between FDI and

local innovative performance we adopt an instrumental variable (IV)

approach based on the “shift-share” methodology associated with Bartik

(1991) and recently popularized by a number of contributions in different

fields (Card, 2007; Moretti, 2010; Overman and Faggio, 2012). To the

best of our knowledge, this methodology has not been adopted so far in

the literature on the impact of FDI, mainly due to the nature of proxies

employed to measure FDI used in the great majority of past studies. This

instrument uses initial shares of employment by division45 in each

province and the average amount of FDI inflows at national level between

2001 and 2006 by division to instrument the amount of FDI that each

province receives during the same time interval. The rationale behind

this instrumental variable builds on the idea that in the absence of area

specific shocks, each province would benefit from a share of national FDI

inflows proportional to its initial share of employment by division taken

as a measure of specialization and calculated in 1991. This further

implies assuming that the location decision of MNEs looks at the

characteristics of the local production system and tends to be skewed

toward areas characterized by a greater potential in terms of backward

and forward linkages, complementarities in production, availability of

trained labour force and local know how (Saliola and Zanfei, 2005). The

45

This is defined in terms of 2-digits NACE classification and data refers to the 1991 Census. Note that the

2-digit dimension has been preferred to more detailed classification in order to account for both broader

sectoral spillovers.

191

instrument is then expected to be significantly and positively correlated

with our regressor of interest due to the traditional stability in the

sectoral specialization of Italian provinces.

More specifically it will takes the following form:

𝐼𝑉_𝐹𝐷𝐼𝑖 = ∑ 𝐸𝑚𝑝𝑙𝑜𝑦𝑚𝑒𝑛𝑡 𝑠ℎ𝑎𝑟𝑒𝑖,1991𝑗

𝑗 × (1 + 𝐹𝐷𝐼𝐼𝑛𝑓𝑙𝑜𝑤𝑠2001−2006𝑗

) (2)

where 𝐹𝐷𝐼2001−2006𝑗

represents the share of FDI inflows in the 2-digits

sector j at national level within the period 2001-2006 and

𝐸𝑚𝑝𝑙𝑜𝑦𝑚𝑒𝑛𝑡 𝑠ℎ𝑎𝑟𝑒𝑖,1991𝑗

is the share of employment in sector j and province

i in 1991. This implies that the flows of FDI at national level by sector are

attributed to each province based on the initial degree of sectoral

specialization.

5.5 Results and discussion

The main results for our specification of interest are reported in Table

5.2.

[Table 5.2 here]

Column 1 presents our baseline specification where the innovative

performance of Italian provinces is regressed on the amount of inputs

devoted to the innovation process, namely investment in R&D and share

of graduates in science and technology. As expected, both innovation

inputs are positively and significantly related to the generation of new

knowledge.

Column 3 includes explicitly the regressor of interest, namely the

amount of FDI as share of provincial GDP, supporting the existence of a

positive and significant correlation at 1% level with the innovative

192

performance of Italian provinces. Further controls are progressively

added in the following columns in order to test for the robustness of our

correlation against the inclusion of potentially relevant variables.

Regressors for population density as proxy of agglomeration, value added

in manufacturing as measure of specialization and long term

unemployment to control for local labour market characteristics are

explicitly included in columns 3, 4 and 5 respectively. Column 5 also

adds a full set of macro-regional dummies to rule out the risk of

unobserved regional characteristics operating at broader geographical

scale. This is a particularly relevant issue in the case of Italy given the

traditional north-south divide within the country.

All the regressors show the expected sign with population density

significantly and positively correlated to local innovative performance and

long term unemployment significantly and negatively associated to

innovation. Interestingly, our control for specialization in manufacturing,

despite entering our regression as significant and positively related to

innovation (Tab.2, Col. 4), becomes gradually less significant once

further controls are included, corroborating our feeling with respect to

the role played by the traditional north-south Italian dichotomy. Finally,

it is worth noting that the inclusion of additional controls lower the

significance of investments in R&D, further supporting the key role of

external capital (complementing the limited financial capacity of the

Italian production system based mainly on small and medium

enterprises) and the availability of an “enabling environment for

innovativeness” (Glaeser et al., 2010) in fostering local innovative

performance.

The regressor of interest, the share of inward FDI, despite the slightly

decreasing magnitude in the coefficient, remains significant at 1% level

and positively correlated to innovation in all specifications.

193

In spite of the evidence in favour of the existence of a robust

correlation between FDI and local innovative performance in the case of

Italian provinces, it has to be acknowledged that our specification,

focusing only on the inflow of FDI, may underestimate the potential

negative effect of foreign disinvestment. The relevance of the latter as key

control for the investigation of the impact of FDI inflows in specific

geographical contexts has been rarely investigated within the existing

literature mainly due to lack of data. Nonetheless, foreign disinvestment

may weaken the local production system and reduce the intensity of

localized knowledge externalities. This is a particularly relevant concern

in the case of Italy where public incentives for the attraction of FDI,

especially in southern regions, have been extensively adopted without

taking properly into account their long run sustainability. In order to

control for this potential negative impact, column 6 includes an

additional regressor for foreign disinvestment. As expected, it enters the

estimation with a negative and significant sign, also contributing to

increase the magnitude of our regressor of interest. This suggests that

disinvestment may have a second order effect in determining the

innovative performance of local areas. This evidence is reasonable in light

of our dependent variable measuring innovation rather than productivity

or growth. The valuable knowledge externalities arising from FDI are

likely to be more relevant in the case of new investment bringing into the

local economy novel distinctive technological capabilities. On the other

side, disinvestment is plausibly affecting more consistently the strengths

of the local production system and weakening the intensity of localized

agglomeration economies. This, in turn, reduces the capability to exploit

the benefits associated to novel information.

In order to try to address the potential bias related to additional

omitted variables and reverse causality we adopt the instrumental

variable approach previously discussed. Results reported in Column 7

(Table 5.2) confirm the positive and highly significant correlation (1%

194

level) of inward FDI with our dependent variable. Despite being not

evident in terms of changes in the significance level of our regressor of

interest, the Hausman test confirms the existence of a substantial bias in

our OLS estimates that justifies the change in the magnitude of our

coefficient. A change of 1% in the share of FDI on provincial GDP

generates a 29% increase in the share of patents application per million

of inhabitants. In the interpretation of this value it should be borne in

mind both the scale of our dependent and independent variables46 and

the nature of our measure of innovation, namely patent applications,

that are likely to be more representative of the innovative performance of

large enterprises rather than small and medium firm. Although few

papers investigate the impact of FDI in the Italian case, the evidence in

favour of a positive effect of FDI correlates with some recent evidences

(Castellani and Zanfei, 2003, 2007). Despite that, any comparison on the

magnitude of the effect remains controversial due to a substantial

difference in the actual variables employed. Most of the existing studies

adopting proxy measures for both FDI and local innovative performance

tend to overlook any further discussion about the actual size of the

economic effect.

The first-stage estimate reported in Table 5.3 confirms the reliability

of our instrument, which is significantly correlated with the

instrumented variable. In addition to that and in compliance with the

econometric literature on weak instruments (Staiger and Stock, 1997;

Stock and Yogo, 2005), the F-statistic for the first-stage is reported in

Table 5.4 showing a value that is generally above both the value of 10

46

Note that in respect to the existing literature our FDI variable reflects the real amount of capital inflows.

An average increase of 1% in the share of FDI over GDP is quantifiable in more than 1 million of Euros

while an increase of 29% in the share of patents over GDP is about 7.46 patents.

195

reported by Staiger and Stock (1997) and the thresholds values defined

by Stock and Yogo (2005).

[Tables 5.3 and 5.4 here]

5.6 Robustness Checks

We start checking the robustness of our results by looking at the

goodness of the instrumental variable approach. Table 5.5 reports our

2SLS estimation progressively eliminating all the controls. The sign and

significance level of our regressor of interest remains unchanged

confirming that its effect on the innovative performance of Italian

provinces is not driven by model specification. This test may also be

taken as indirect evidence supporting the validity of the exclusion

restrictions.

[Insert Table 5.5 here]

Finally, in order to provide further support to the instrumental

variable approach, the reduced form equation is estimated by means of

OLS regression of our dependent variable on the instrument and

exogenous controls (Table 6). As shown by Angrist and Krueger (2001),

although being poorly informative with respect to the real magnitude of

the coefficient, the reduced form can be used as additional test to

determine the sign of the coefficient of interest. The estimation of the

reduced form equation confirms that FDI is a positive and relevant

determinant of innovation in Italian provinces.

[Insert Table 5.6 here]

196

The estimation performed in this paper demonstrates to be robust to

the inclusion of additional significant regressors and to the correction of

the potential bias associated with the endogeneity of the regressor of

interest. The instrumental variable approach discussed has showed to be

strongly correlated to the instrumented variable and not affected by

issues related to the specification of the model. Nevertheless, there is still

the possibility that our instrument is correlated with other variables not

explicitly taken into account in our regression. According to the existing

literature on the impact of FDI, it is reasonable to assume that our

instrument for FDI is correlated with a negative competition effect

provoking the exit of local firms from the market. Despite being

acknowledged by many existing studies, this issue is rarely explicitly

addressed in the literature mainly due to lack of data. Nonetheless, a

negative competition effect due to the entry of MNEs with superior

technological, managerial and organizational skills (Cantwell and

Iammarino, 2003) crowding out local firms may impact the structure of

the local production system weakening local innovative potentials. To

control for this specific aspect, our instrument has been regressed over

the provincial share of domestic companies in liquidation. Results

reported in Table 5.7 rule out any doubts regarding a potential

systematic correlation with our instrument.

[Table 5.7 here]

Finally, we perform a further check for the robustness of our results

by considering a dependent variable that is more commonly employed in

the existing literature, namely, labour productivity47. Results shown in

Table 5.8 confirm that the key intuition does not change when a more

47

This is measured as the value added in manufacturing per unit of labour. Data are available at the

provincial level for 2001-2006 and comes from ISTAT. While labour productivity is widely used in the

literature, TFP has to be conceptually preferred. However, data on TFP is not available at NUTS-3 level.

197

traditional analytical framework looking at the spillover effects of FDI on

a measure of productivity is taken into consideration. FDI still exhibits a

significant and positive sign both in the OLS and IV specification.

Therefore, this result suggests that the evidence in favour of FDI-induced

externalities persists also within an empirical setting that is more

coherent with previous studies.

[Table 5.8 here]

The estimation of the impact of FDI on local innovative performance

seems to be robust to a number of checks, encompassing the inclusion of

additional controls and the implementation of the 2SLS estimation to

address the endogeneity of the regressor of interest. FDI proved to be a

significant determinant of local innovative performance by enriching the

knowledge base of Italian provinces and generating valuable positive

knowledge externalities.

5.7 Conclusion

In the last few decades the attraction of FDI has been placed at the

core of the policy agenda in both developed and developing countries.

This centrality in the political debate is supported by the belief that the

attraction of external resources could benefit recipient economies thanks

to knowledge externalities arising from the localization of affiliates of

MNEs endowed with superior technological, managerial and

organizational skills.

However, existing academic literature suggests that local economic

conditions are a crucial pre-requisite for valuable knowledge externalities

to be successfully captured by local production systems and transformed

in innovation.

198

So far there is weak consensus on whether knowledge externalities

associated to FDI benefit systematically the economic and innovative

performance of host locations. Such an inconclusiveness of the existing

empirical literature is due to a number of flaws.

A first point concerns measurement issues associated to the adoption

of proxies for both FDI and innovative performance. Traditionally, FDI is

indirectly measured by indicators of foreign presence such as the share

of employment in foreign firms or the number of foreign firms. These

variables do not account for the actual size of foreign capital mixing up

relevant financial investment with minor flows. A second concern regards

the endogeneity related to the estimation of the causal impact of FDI.

While early literature generally focuses on the correlation between FDI

and outcome variables, more recently scholars have paid deeper

attention to these sources of biasedness. However, there are still few

attempts to track consistently this issue and more work is needed in this

direction.

This paper aims at contributing to the existing debate with new

evidence and attempts to address both the above mentioned problems.

Firstly, we adopt a direct measure of FDI consisting of the real amount of

capital flow in Italian provinces. Secondly, we try to tackle endogeneity

concerns through IV methodologies. In our empirical exercise, we find

that FDI contributes significantly to the patenting activities of Italian

provinces over the period 2001-2006. This finding correlates with similar

evidence provided by some previous empirical studies.

Beside of this, it is worth noting that our investigation focuses on the

gross impact of FDI without disentangling the channels through which

knowledge externalities affect local economies. It remains in our future

research agenda the development of a more detailed investigation of the

mechanisms through which knowledge spills over space. This further

direction for research will be possible along with improvements in the

quality of data. An additional limitation of this final chapter also

199

concerns the cross-sectional nature of our data for estimation purposes:

not including province-level fixed effects, in fact, might introduce an

estimation bias even in the IV equation, given that unobserved drivers of

innovation could be correlated with FDI. Although this represents a

potential issue, we are confident that the impact of this unobserved

source of heterogeneity is very limited when the exogenous instrument is

adopted. Therefore, this represents another area for improvement in our

future research agenda.

In terms of policy considerations, results suggest that FDI can play

an important role in fostering local innovative outcomes. Therefore, local

economies should consider external sources of knowledge as a

complement to internal generation (Bathelt et al., 2004). This is even

more relevant considering that a core local input of innovation such as

private R&D seems to be less important than expected in our empirical

exercise probably due to the structure of the Italian production system

based on a great share of small and medium enterprises with a

reasonably limited financial capacity. Our results are also important in

light of the well-known historically poor amount of FDI that Italy receives

annually as compared to other large European countries, such as UK,

Germany and France. Italy has never adopted any structured policy

oriented to the attraction of FDI. The empirical evidence provided

suggests that creating an actual policy of FDI attraction that stimulates

foreign investors might be a valuable policy option. Clearly, while our

results suggest that FDI can be beneficial per se, we are obviously

cautious in arguing that Italian provinces should attract FDI

irrespectively of local strength and weaknesses in terms of specialization

of labour force and specialization and competencies of local firms.

Indeed, the specific profile of local economies have been shown to play a

strong role in shaping the effectiveness of knowledge externalities arising

from FDI as demonstrated by the relevance of additional localized drivers

200

of knowledge generation such as human capital and agglomeration

economies.

201

Figure 5.1: Share of Inward FDI per Macro-Region

Source: Authors’ elaborations on Bank of Italy data.

0.5

11.5

FD

I share

2001 2002 2003 2004 2005 2006Year

Inward FDI - National

0.5

11.5

FD

I share

2001 2002 2003 2004 2005 2006Year

Inward FDI - North0

.51

1.5

FD

I share

2001 2002 2003 2004 2005 2006Year

Inward FDI - Centre

0.5

11.5

FD

I share

2001 2002 2003 2004 2005 2006Year

Inward FDI - South & Islands

202

Table 5.1: Variables List

Variable Definition Source Geography Time

Patents Applications to EPO

(by applicants) OECD Provincial

2001-

2006

Private R&D Share of expenditure for private R&D on

GDP

ISTAT Regional 2001-2006

Graduates Share of graduates in

science and technology on population

ISTAT Regional 2001-2006

FDI Millions in national currency

Bank of Italy Provincial 2001-2006

Population Density

Population on

provincial surface ISTAT Provincial

2001-

2006

Employment in Manufacturing

Share of employment in manufacturing on

total employment

OECD Provincial 2001-

2006

Long Term Unemployment

Share of long term unemployed on

population

ISTAT Regional 2001-2006

Foreign Disinvestment

Millions in national currency

Bank of Italy Provincial 2001-2006

Firms in Liquidation

Share of firms in liquidation on total number of firms

Unioncamere Provincial 2001-

2006

Labour Productivity

Value added in manufacturing per

unit of labour

ISTAT Provincial 2001-2006

Notes: a) Patents, FDI and Foreign Disinvestment variables are weighted by provincial GDP, measured in millions of national currency (source: OECD); b) all variables are averaged over the period 2001-2006 and enter regressions in log form.

203

Table 5.2: Inward FDI and Local Innovative Performance

(1) (2) (3) (4) (5) (6) (7)

Dep.Var. Patents OLS OLS OLS OLS OLS OLS 2SLS

Private R&D 0.476*** 0.335** 0.340** 0.188 0.0544 0.0211 -0.0148

(0.154) (0.167) (0.159) (0.161) (0.136) (0.148) (0.172)

Graduates 0.633** 0.707** 0.674** 0.669** 0.427*** 0.468*** 0.719**

(0.309) (0.300) (0.294) (0.301) (0.154) (0.166) (0.297)

FDI

0.137*** 0.134*** 0.0993*** 0.0675*** 0.0782*** 0.296***

(0.0373) (0.0354) (0.0319) (0.0251) (0.0241) (0.0541)

Population Density

0.311* 0.327* 0.420*** 0.449*** 0.383**

(0.182) (0.177) (0.156) (0.155) (0.173)

Employment in

Manufacturing

1.170*** 0.511* 0.505* 0.0385

(0.268) (0.263) (0.259) (0.308)

Long Term Unemployment

-0.578*** -0.486*** -0.397***

(0.0805) (0.142) (0.153)

Foreign Disinvestment

-0.0467** -0.0728**

(0.0185) (0.0321)

Constant -7.592*** -6.956*** -8.561*** -7.560*** -8.125*** -8.646*** -8.341***

(0.806) (0.800) (1.377) (1.420) (1.183) (1.189) (1.452)

Macro-Regional dummies NO NO NO NO NO YES YES

Observations 103 103 103 103 103 103 103

R-squared 0.380 0.489 0.510 0.579 0.684 0.707 0.479

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

204

Table 5.3: First Stage Regression

(1)

Dep.Var.: FDI Inflows OLS

Private R&D 0.0847

(0.4254)

Graduates -1.1870

(0.8479)

Population Density 0.0277

(0.4472)

Employment in

Manufacturing 1.9674**

(0.9414)

Long Term Unemployment -0.4595

(0.4398)

Foreign Disinvestment 0.0798

(0.1024)

IV FDI 4.9374***

(1.2091)

Constant -1.1380

(3.5854)

Macro-Regional Dummies YES

Observations 103

R-squared 0.314

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

Table 5.4: First Stage Statistics

Variable F(1, 93) P-Value

IV FDI 16.67 0.000

205

Table 5.5: Model Specification

(1) (2) (3) (4) (5) (6)

Dep.Var.: Patents 2SLS 2SLS 2SLS 2SLS 2SLS 2SLS

FDI 0.296*** 0.300*** 0.233*** 0.260*** 0.275*** 0.305***

(0.0541) (0.0583) (0.0423) (0.0379) (0.0459) (0.0424)

Private R&D -0.0148 -0.00440 0.00445 0.103 0.195 0.162

(0.172) (0.149) (0.152) (0.175) (0.177) (0.192)

Graduates 0.719** 0.686*** 0.571** 0.754** 0.754** 0.799**

(0.297) (0.259) (0.231) (0.324) (0.318) (0.340)

Population Density 0.383** 0.382** 0.362** 0.291 0.283

(0.173) (0.172) (0.171) (0.180) (0.186)

Employment in Manufacturing 0.0385 0.0209 0.186 0.669***

(0.308) (0.315) (0.276) (0.256)

Long Term Unemployment -0.397*** -0.389*** -0.432***

(0.153) (0.111) (0.112)

Foreign Disinvestment -0.0728** -0.0731**

(0.0321) (0.0324)

Constant -8.341*** -8.269*** -7.591*** -7.162*** -7.762*** -6.172***

(1.452) (1.480) (1.372) (1.513) (1.467) (0.871)

Macro-Regional dummies YES NO NO NO NO NO

Observations 103 103 103 103 103 103

R-squared 0.479 0.468 0.546 0.443 0.395 0.324

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

206

Table 5.6: Reduced Form Equation

(1)

Dep.Var.: Patents OLS

IV FDI 1.461***

(0.384)

Private R&D 0.0103

(0.143)

Graduates 0.368**

(0.145)

Population Density 0.391**

(0.160)

Employment in Manufacturing 0.621**

(0.254)

Long Term Unemployment -0.533***

(0.142)

Foreign Disinvestment -0.0492**

(0.0188)

Constant -8.677***

(1.169)

Macro-Regional dummies YES

Observations 103

R-squared 0.715

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

Table 5.7: Market Exit

(1)

Dep.Var.: IV FDI OLS

Firms in Liquidation 0.157

(0.0987)

Constant 0.654*

(0.355)

Observations 103

R-squared 0.115

Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

207

Table 5.8: Inward FDI and Labour Productivity

(1) (2) Dep.Var: Labour

Productivity OLS 2SLS

Private R&D 0.0046 0.0031

(0.0198) (0.0201)

Graduates 0.0355 0.0458

(0.0348) (0.0396)

FDI 0.0145*** 0.0235**

(0.0053) (0.0103)

Constant 10.78*** 10.79***

(0.177) (0.174)

Controls YES YES

Macro-Regional dummies YES YES

Observations 103 103

R-squared 0.407 0.385

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

208

Bibliography

Acemoglu D. and Johnson S., 2005. Unbundling institutions. Journal of

Political Economy 113, 949-995.

Acemoglu D., Johnson S. and Robinson J. A., 2001. The colonial origins of

comparative development: An empirical investigation. American Economic

Review 91, 1369-1401.

Acemoglu D., Johnson S. and Robinson J. A., 2005. Institutions as a

fundamental cause of long-run growth. In Aghion P. and Durlauf S. N.

(eds.), Handbook of Economic Growth, Volume 1A. Elsevier.

Alcácer J. 2006. Location choices across the value chain: how activity and

capability influence colocation. Management Science 52(10), 1457–1471.

Almeida P., 1996. Knowledge sourcing by foreign multinationals: Patent

citation analysis in the US semiconductor industry. Strategic

Management Journal 17, 155-165.

Alsan M., Bloom D. E. and Canning D., 2006. The effect of population health

on foreign direct investment inflows to low- and middle-income countries.

World Development 34, 613-630.

Altomonte C., 2000. Economic determinants and institutional framework:

FDI in economies in transition. Transnational Corporations 9, 75-106.

Amighini A. and Rabellotti R., 2006. How do Italian footwear industrial

districts face globalization? European Planning Studies 14, 485-502.

Andrade G., Mitchell M. and Stafford E., 2001. New evidence and

perspectives on mergers. Journal of Economic Perspectives 15, 103-120.

209

Angrist J.D. and Krueger A. B., 2001. Instrumental variables and the search

for identification: From supply and demand to natural experiment.

Journal of Economic Perspectives 15(4): 69-85.

Antrás P., 2003. Firms, contracts and trade structure. Quarterly Journal of

Economics 118, 1375-1418.

Antrás P. and Helpman E., 2004. Global sourcing. Journal of Political

Economy 112, 552-580.

Antrás P. and Helpman E., 2008. Contractual frictions and global sourcing.

In Helpman E., Marin D. and Verdier T. (eds), The Organization of Firms

in a Global Economy, Harvard University Press.

Arnold J. M. and Javorcik B. S., 2009. Gifted kids or pushy parents? Foreign

direct investment and plant productivity in Indonesia. Journal of

International Economics 79, 42-53.

Asiedu E., 2002. On the determinants of foreign direct investment to

developing countries: Is Africa different? World Development 30, 107-119.

Asiedu E., 2006. Foreign direct investment in Africa: The role of natural

resources, market size, government policy, institutions and political

instability. The World Economy 29, 63-77.

Baltagi B.H., 2005. Econometric Analysis of Panel Data, John Wiley and

Sons.

Bank of Italy, 2000. Le economie del Mediterraneo. Banca d’Italia, Rome

Balsvik R. and Haller S., 2010. Picking “lemons” or picking “cherries”?

Domestic and foreign acquisitions in Norwegian manufacturing. The

Scandinavian Journal of Economics 112, 361-387.

Barrel R. and Pain N., 1999. The growth of foreign direct investment in

Europe. In Barrell R. and Pain N. (eds), Innovation, Investment and the

Diffusion of Technology in Europe. Cambridge University Press, UK.

210

Bartik T.J., 1991. Who Benefits from State and Local Economic Development

Policies? Kalamazoo: Upjohn Institute Press.

Basile R., Castellani D. and Zanfei A., 2008. Location choices of

multinational firms in Europe: The role of EU cohesion policy. Journal of

International Economics 74, 328-340.

Bathelt H., Malmberg A. and Maskell P., 2004. Clusters and knowledge:

Local buzz, global pipelines and the process of knowledge creation.

Progress in Human Geography 28, 31-56.

Bellak C. and Leibrecht M., 2009. Do low corporate income tax rates attract

FDI? – Evidence from Central- and East European countries. Applied

Economics 41, 2691-2703.

Bellak C., Leibrecht M. and Riedl A., 2008. Labour cost and FDI flows into

Central and Eastern European Countries: A survey of the literature and

empirical evidence. Structural change and Economic Dynamics 19, 17-37.

Bénassy Quéré A., Coupet M. and Mayer T., 2007. Institutional

determinants of foreign direct investment. The World Economy 30, 764-

782.

Benfratello L. and Sembenelli A., 2006. Foreign ownership and productivity:

Is the direction of causality so obvious? International Journal of Industrial

Organization 24, 733-751.

Besley T., 1995. Property rights and investment incentives: Theory and

evidence from Ghana. Journal of Political Economy 103, 903-937.

Beugelsdijk, S., and Mudambi, R., 2013. MNEs as border-crossing multi-

location enterprises: The role of discontinuities in geographic space.

Journal of International Business Studies, 44, 413-426.

Bevan A. and Estrin S., 2004. The determinants of foreign direct investment

into European transition economies. Journal of Comparative Economics

32,775-787.

211

Bevan A., Estrin S. and Meyer K., 2004. Foreign investment location and

institutional development in transition economies. International Business

Review 13, 43-64.

Billington N., 1999. The location of foreign direct investment: An empirical

analysis. Applied Economics 31, 65-76.

Bitzer J., Geishecker I. and Görg H., 2008. Productivity spillover through

vertical linkages: Evidence from 17 OECD countries. Economics Letters

99, 328-331.

Bjorvatn K., 2004. Economic integration and the profitability of cross-border

mergers and acquisitions. European Economic Review 48, 1211-1226.

Blalock G., 2001. Technology from foreign direct investment: Strategic

transfer through supply chains. Paper presented at the Empirical

Investigations in International Trade Conference at Purdue University,

November 9–11, 2001 (part of doctoral research at Haas School of

Business, University of California, Berkeley).

Blalock G. and Gertler P.J., 2008. Welfare gains from foreign direct

investment through technology transfer to local suppliers. Journal of

International Economics 74, 402-421.

Blomström M., 1989. Foreign Investment and Spillovers. Routledge, London

and New York.

Blomström M. and Kokko A., 1998. Multinational corporations and

spillovers. Journal of Economic Surveys 12 (2), 247-277.

Blomström M. and Persson H., 1983. Foreign investment and spillover

efficiency in an underdeveloped economy: Evidence from the Mexican

manufacturing industry. World Development 11 (6), 493-501.

Blonigen B., Fontagné L., Sly N. and Toubal F., 2014. Cherries for sale: The

incidence and timing of cross-border M&A. Journal of International

Economics 94, 341-357.

212

Boeri T. and Brücker H., 1999. Eastern Enlargement and EU-Labour

Markets: Perceptions, Challenges and Opportunities. IZA Discussion

Paper No. 256.

Boschma R. and Capone G., 2013. Relatedness and diversification in EU

and ENP countries. SEARCH Working Paper WP2.4.

Boschma R.A. and Wenting R., 2007. The spatial evolution of the British

automobile industry: Does location matter? Industrial and Corporate

Change 16 (2), 213-238.

Brakman S., Garretsen H., van Marrewijk C. and van Witteloostuijn A. 2013.

Cross border mergers and acquisition activity and revealed comparative

advantage in manufacturing industries. Journal of Economics &

Management Strategy 22, 28-57.

Breinlich H., 2008. Trade liberalization and industrial restructuring though

mergers. Journal of International Economics 76, 254-266.

Brenton P., Di Mauro F. and Lücke M., 1999. Economic Integration and FDI:

An Empirical Analysis of Foreign Investment in the EU and in Central

and Eastern Europe. Empirica 26, 95-121.

Buckley P. J. and Casson M. C., 1976. The Future of Multinational

Enterprise. London: Macmillan.

Cainelli G., Mancinelli S. and Mazzanti M., 2005. Social capital, R&D and

industrial District. Feem Working papers n. 744584.

Campa J. and Kedia S., 2002. Explaining the diversification discount.

Journal of Finance 57, 1731-1762.

Campos N. and Kinoshita Y., 2003. Why does FDI go where it goes? New

evidence from the transition economies. IMF working Paper.

Cantwell J., 1989. Technological innovation and Multinational Corporations.

Basil Blackwell: Oxford, UK.

213

Cantwell J., 2009. Location and the multinational enterprise. Journal of

International Business Studies 40, 35-41.

Cantwell J. and Iammarino S., 2003. Multinational Corporations and

European Regional System of Innovation. London and New York:

Routledge.

Cantwell J. and Piscitello L., 2005. Recent location of foreign-owned

research and development activities by large multinational corporations

in the European regions: The role of spillovers and externalities. Regional

Studies 39, 1-16.

Cantwell, J. and Santangelo, G.D.,1999. The frontier of international

technology networks: sourcing abroad the most highly tacit capabilities.

Information Economics and Policy 11, 101‒123.

Card D., 2007. How immigration affects US cities. CReAM Discussion Paper

11/07.

Carabelli A., Hirsch G. and Rabellotti R., 2009, Italian SMEs and Industrial

Districts on the move: where are they going?, European and Planning

Studies, 17, 1, 19-41

Carstensen K. and Toubal F., 2004. Foreign direct investment in Central

and Eastern European countries: a dynamic panel analysis. Journal of

Comparative Economics 32, 3-22.

Casson M. C., 1979. Alternatives to the Multinational Enterprise. London:

Macmillan.

Castellani D. and Giovannetti G., 2010. Productivity and the international

firm: Dissecting heterogeneity. Journal of Economic Policy Reform 13, 25-

42.

Castellani D. and Zanfei A., 2003. Technology gaps, absorptive capacity and

the impact of inward investments on productivity of European firms.

Economics of Innovation and New Technology 12, 555-576.

214

Castellani D. and Zanfei A., 2006. Multinational Firms, Innovation and

Productivity. Edward Elgar, Cheltenham, UK – Northampton, MA, USA.

Castellani D. and Zanfei A., 2007. Multinational Companies and Productivity

Spillovers: Is There a Specification Error? Applied Economic Letters 14,

1047-1051.

Coeurdacier N., De Santis R. and Aviat A., 2009. Cross-border mergers and

acquisitions and European integration. Economic Policy 24, 55-106.

Cornett M., Marcus A. and Tehranian H., 2008. Corporate governance and

pay-for-performance: The impact of earnings management. Journal of

Financial Economics 87, 357-373.

Criscuolo C. and Martin R., 2009. Multinationals and US productivity

leadership: Evidence from Great Britain. The Review of Economics and

Statistics 91, 263-281.

Caves R., 1974. Multinational firms, competition, and productivity in host-

country markets. Economica 41, 176-193.

Caves R., 1996. Economic Analysis and the Multinational Corporation (2nd

Ed.). Cambridge University Press: Cambridge, UK.

Chen W., 2011. The effect of investor origin on firm performance: Domestic

and foreign direct investment in the United States. Journal of

International Economics 83, 219-228.

Chen H. and Chen T.J., 1998. Network linkages and location choice in

foreign direct investmen. Journal of International Business Studies 29,

445-467.

Chen M. and Moore M., 2010. Location decision of heterogeneous

multinational firms. Journal of International Economics 80, 188-199.

Cheng S., 2008. Location decision variations of Japanese investors in China.

The Review of Regional Studies 38, 395-415.

215

Cheung K. and Lin P., 2004. Spillover effects of FDI on innovation in China:

Evidence from the provincial data. China Economic Review 15, 25-44.

Cieslik, A. and Hagemejer, J., 2009. Assessing the Impact of the EU-

sponsored Trade Liberalization in the MENA Countries, Journal of

Economic Integration 24, 343-368.

Commission of the European Communities, 2004. European Neighbourhood

Policy Strategy Paper. COM(2004) 373 final, Brussels.

COM, 2004. European Neighbourhood Policy Startegy Paper. Commission of

the European Communities.

Cooke W.N., 1997. The influence of industrial relations factors on US foreign

direct investment abroad. Industrial and Labor Relation Review 51, 3-17.

Coughlin C. C., Terza J. V and Arromdee V., 1991. State characteristics and

the location of foreign direct investment within the United States. The

Review of Economics and Statistics 73, 675-683.

Crescenzi R., Gagliardi L. and Percoco M., 2013. Social capital and the

innovative performance of Italian provinces. Environment and Planning A

45, 908-929.

Crescenzi R., Pietrobelli C. and Rabellotti R., 2014. Innovation drivers, value

chains and the geography of multinational corporations in Europe.

Journal of Economic Geography, 14, 1053-1086.

Crescenzi R., Rodríguez-Pose A. and Storper M., 2007. The territorial

dynamics of innovation: a Europe–United States comparative analysis.

Journal of Economic Geography 7, 673–709.

Crespo N. and Fontoura M.P., 2007. Determinant factors of FDI spillovers –

What do we really know? World Development 35 (3), 410-425.

Crespo N., Fontoura M.P. and Proença I., 2009. FDI spillovers at the regional

level: Evidence from Portugal. Papers in Regional Science 88 (3), 591-607.

216

Criscuolo C., Haskel J.E. and Slaughter M.J., 2010. Global engagement and

the innovation activities of firms. International Journal of Industrial

Organization 28, 191-202.

Crozet M., Mayer T. and Mucchielli J., 2004. How do firms agglomerate? A

study of FDI in France. Regional Science and Urban Economics 34, 27-54.

Daniele V. and Marani U., 2011. Organized crime, the quality of local

institutions and FDI in Italy: A panel data analysis European Journal of

Political Economy 27, 132–142

Daude, C. and Stein, E., 2007. The quality of institutions and foreign direct

investment. Economics & Politics 19, 317–344.

Defever F., 2006. Functional fragmentation and the location of multinational

firms in the enlarged Europe. Regional Science and Urban Economics 36,

658-677.

Defever, 2012. The spatial organization of multinational firms. Canadian

Journal of Economics 45, 672-697.

Deichman J., Karidis S. and Sayek S., 2003. Foreign direct investment in

Turkey: Regional determinants. Applied Economics 35, 1767-1778.

Devereux M. P. and Griffith R., 1998. Taxes and the location of production:

Evidence from a panel of US multinationals. Journal of Public Economics

68, 335-367.

Devereux M. P., Griffith R. and Simpson H., 2007. Firm location decisions,

regional grants and agglomeration externalities. Journal of Public

Economics 91, 413-435.

Dewenter K. and Malatesta P., 2001. State-owned and privately-owned

firms: An empirical analysis of profitability, leverage and labor intensity.

American Economic Review 91, 320-334.

217

Dewit G., Görg H. and Montagna C., 2003. Should I stay or should I go? A

note on employment protection, domestic anchorage, and FDI. IZA

Discussion Paper 845.

Díez F. and Spearot A., 2014. Core competencies, matching and the

structure of foreign direct investment. Canadian Journal of

Economics/Revue canadienne d’économie 47,813-855.

Dicken P., 2007. Global Shift: Mapping the Changing Contours of the World

Economy. Fifth edition, Sage, London.

Djankov S., McLiesh C. and Ramalho R.M., 2006. Regulation and growth.

Economic Letters 92, 395-401.

Disdier A. and Mayer T., 2004. How different is Eastern Europe? Structure

and determinants of location choices by French firms in Eastern and

Western Europe. Journal of Comparative Economics 32, 280-296.

Driffield N., 2006. On the search for spillovers from foreign direct investment

(FDI) with spatial dependency. Regional Studies 40 (1), 107-119.

Du J., Lu Y. and Tao Z., 2008. Economic institutions and FDI location

choice: Evidence from US multinationals in China. Journal of

Comparative Economics 36, 412-429.

Dunford M., 2006. Industrial districts, magic circles, and the restructuring

of the Italian textiles and clothing chain. Economic Geography 82, 27-59.

Dunning J.H., 1977. Trade, location of economic activity. Proceedings of a

Nobel Symposium held at Stockholm, London: McMillan, 395-418.

Dunning J.H., 1980. Toward an eclectic theory of international production:

Some empirical tests. Journal of International Business Studies 11(1), 9-

31.

Dunning J.H., 1988. The eclectic paradigm of international production: in

defence of the eclectic theory. Journal of International Business Studies

19, 1-31.

218

Dunning J.H., 2009. Location and the multinational enterprise: John

Dunning’s thoughts on receiving the Journal of International Business

Studies 1008 Decade Award. Journal of International Business Studies

40, 20-34.

Dunning J. and Lundan S., 2008. Multinational Enterprises and the Global

Economy (2nd Ed.). Edward Elgar.

Duranton G., 2006. Human Capital Externalities in Cities: Identifycation

and Policy Issues. In Arnott R. and McMillen D. (Eds.) A companion to

Urban Economics. Oxford: Blackwell Publishing Ltd.

Ekholm K., Forslid R. and Markusen J. R., 2007. Export-platform foreign

direct investment. Journal of the European Economic Association 5, 776-

795.

Erel I. Liao R. and Weisbach M., 2012. Determinants of cross-border

mergers and acquisitions. Journal of Finance 57, 1045-1082.

Ernst D. and Kim L., 2002. Global production networks, knowledge

diffusion, and local capability formation. Research Policy 31, 1417-1429.

European Commission, 2014, Implementation of the European

Neighbourhood Policy in 2013 - Regional report: A Partnership for

Democracy and Shared Prosperity with the Southern Mediterranean

Partners, Joint Staff Working Document, Brussels, 27.3.2014 SWD(2014)

100 final

Fabry N. and Zeghni S., 2002. Foreign direct investment in Russia: How the

investment climate matters. Communist and Post-Communist Studies 35,

289-303.

Faggio G. and Overman H.G., 2012. The effect of public employment on local

labour markets. SERC DP 111.

219

Feliciano Z. and Lipsey R., 2015. Foreign entry into US manufacturing by

takeovers and the creation of new firms. Eastern Economic Journal,

doi:10.1057/eej.2015.4.

Fich E., Cai J. and Tran A., 2011. Stock option grants to target CEOs during

private merger negotiations. Journal of Financial Economics 101, 413-

430.

Flowers E. B., 1976. Oligopolistic reactions in European and Canadian

direct investment in the United States. Journal of International Business

Studies 7.

Fosfuri A., Motta M. and Rønde T., 2001. Foreign direct investment and

spillovers through workers’ mobility. Journal of International Economics

53, 205-222.

Foster L., Haltiwanger J. and Syverson C., 2008. Reallocation, firm turnover,

and efficiency: Selection on productivity or profitability?" American

Economic Review 98, 394-425.

Fuller C., 2005. Corporate Repeat Investment and Regional Institutional

Capacity. The Case of After-care Services in Wales. European Urban and

Regional Studies 12, 5–21.

Gagliardi L., 2015. Does Skilled Migration Foster Innovative Performance?

Evidence from British Local Areas. Papers in Regional Science,

forthcoming.

Gal P., 2013. Measuring total factor productivity at the firm level using

OECD-Orbis. OECD Economics Department Working Paper No. 1049.

Giannetti M. and Ongena S., 2012. ‘Lending by example’: Direct and indirect

effects of foreign banks in emerging markets. Journal of International

Economics 86, 167-180.

Glaeser E., Kallal H., Scheinkman J. and Shleifer A., 1992. Growth in cities.

Journal of Political Economy 100,1126-1152.

220

Glaeser E.L., Rosenthal S.S. and Strange W.C., 2010. Urban economics and

entrepreneurship. Journal of Urban Economics, 67(1), 1-14.

Glaeser E., Scheinkman J. and Shleifer A., 2003. The injustice of inequality.

Journal of Monetary Economics 50, 199-222.

Glaeser E. and Shleifer A. 2002. Legal origins. Quarterly Journal of

Economics 116, 853-900.

Globerman S., 1979. Foreign direct investment and ‘spillover’ efficiency

benefits in Canadian manufacturing industries. The Canadian Journal of

Economics 12 (1), 42-56.

Globerman S. and Shapiro, D., 2002. Global foreign direct investment flows:

the role of governance infrastructure. World Development 30, 1899–1919.

Goldstein M. and Udry C., 2008. The profits of power: Land rights and

agricultural investment in Ghana. Journal of Political Economy 116, 981-

1022.

Görg H. and Greenaway D., 2004. Much ado about nothing? Do domestic

firms really benefit from foreign direct investment? The World Bank

Research Observer 19 (2), 171-197.

Görg H. and Strobl E., 2005. Spillovers from foreign firms through worker

mobility: An empirical investigation. Scandinavian Journal of Economics

107 (4), 693-709.

Grossman G. and Helpman E., 1991. Trade, knowledge spillovers, and

growth. European Economic Review 35, 517-526.

Grossman G. and Helpman E., 2002. Integration versus outsourcing in

industry equilibrium. Quarterly Journal of Economics 117, 85-120.

Guadalupe M., Kuzmina O. and Thomas C., 2012. Innovation and foreign

ownership. American Economic Review 102, 3594-3627.

221

Guimarães P., Figueiredo O. and Woodward D., 2000. Agglomeration and

the location of foreign direct investment in Portugal. Journal of Urban

Economics 47, 115-135.

Guimarães P., Figueiredo O. and Woodward D., 2003. A Tractable Approach

to the Firm Location Decision Problem. The Review of Economics and

Statistics 85, 201-204.

Guimarães P., Figueiredo O. and Woodward D., 2004. Industrial Location

Modeling: Extending the Random Utility Framework. Journal of Regional

Science 44, 1-20.

Hale G. and Long, C., 2006. What determines technological spillovers of

foreign direct investment: Evidence from China. Center Discussion Paper

934, Yale University.

Hall R.E. and Jones C.I., 1999. Why do some countries produce so much

more output per worker than others? Quarterly Journal of Economics

114, 83-116.

Haucap J., Wey C. and Barmbold J.F., 1997. Location choice as a signal for

product quality: The economics of ‘Made in Germany’. Journal of

Institutional and Theoretical Economics 153, 510-531.

Harris R., 2009. Spillover and backward linkage effects of FDI: Empirical

evidence for the UK. Spatial Economics Research Centre (SERC)

Discussion Paper 16.

Harris R., Stewart J., Guilkey D and Carleton W., 1982. Characteristics of

acquired firms: Fixed and random coefficients probit analyses. Southern

Economic Journal 49, 164-184.

Haskel J.E., Pereira S.C. and Slaughter M.J., 2007. Does inward foreign

direct investment boost the productivity of domestic firms? The Review of

Economics and Statistics 89 (3), 482-496.

222

Head K. and Mayer T., 2004. Market potential and the location of Japanese

investment in the European Union. Review of Economics and Statistics

86, 959-972.

Head K. and Ries J., 2008. FDI as an outcome of the market for corporate

control: Theory and evidence. Journal of International Economics 74, 2-

20.

Head K., Ries J. and Swenson D., 1995. Agglomeration benefits and location

choice: Evidence from Japanese manufacturing plants. Journal of

International Economics 38 (3-4), 223-247.

Head K., Ries J. and Swenson D., 1999. Attracting foreign manufacturing:

Investment promotion and agglomeration. Regional Science and Urban

Economics 29, 197-218.

Hellman J., 1998. Winners take all: The politics of partial reform in

postcommunist transitions. World Politics 50, 203-234.

Hellman J., Jones G. and Kaufmann D., 2000. Seize the state, seize the day:

state capture, corruption, and influence in transition. World Bank

Working Paper No.2444.

Henisz W. J., 2000. The institutional environment for multinational firms.

The Journal of Law, Economics and Organization 16, 334-364.

Hennart J.F., 1982. A Theory of Multinational Enterprise. Ann Arbor, MI:

University of Michigan Press.

Hilber C. A. and Voicu I., 2010. Agglomeration economics and the location of

foreign direct investment: Empirical Evidence from Romania. Regional

Studies 44, 355-371.

Hisarciklilar M., Kayam S. and Kayalica O., 2006. Locational drivers of FDI

in MENA countries: A spatial attempt. MPRA Working Paper No.2085.

Horn B. and Persson L., 2001. The equilibrium ownership of an

international oligopoly. Journal of International Economics 53, 307-333.

223

Huang L., Liu X. and Xu L., 2012. Regional innovation and spillover effects

of foreign direct investment in China: A threshold approach. Regional

Studies 46, 583-596.

Hymer S.H., 1976. The International Operations of National Firms. The

Massachusetts Institute of Technology Press (originally presented as the

author’s thesis, MIT 1960).

Iammarino S. and Marinelli E., 2011. Is the grass greener on the other side

of the fence? Graduate regional mobility and job satisfaction in Italy.

Environment and Planning A 43(11), 2761-2777.

Iammarino S. and McCann P., 2013. Multinationals and Economic

Geography: Location, Technology and Innovation. Edward Elgar,

Cheltenham UK and Northampton (MA) USA.

Iammarino S. and McCann P., (forthcoming 2015). MNE innovation

networks and the role of cities, in D. Archibugi and A. Filippetti (Eds),

The Handbook of Global Science, Technology and Innovation. Wiley-

Blackwell, Oxford.

Jaffe A. B., Trajtenberg M., and Henderson R., 1993. Geographic localization

of knowledge spillovers as evidenced by patent citations. The Quarterly

Journal of Economics 108, 577-598.

Janicki H. and Wunnava P., 2004. Determinants of foreign direct

investment: empirical evidence from EU accession countries candidates.

Applied Economics 36, 505-509.

Javorcik B.S., 2004. Does foreign direct investment increase the productivity

of domestic firms? In search of spillovers through backward linkages.

The American Economic Review 94 (3), 605-627.

Javorcik, B.S. and Spatareanu M., 2005. Do foreign investors care about

labour market regulations? Review of World Economics 141, 375-403.

224

Javorcik B.S. and Spatareanu M., 2008. To share or not to share: does local

participation matter for spillovers from foreign direct investment? Journal

of Development Economics 85, 194-217.

Javorcik B.S. and Spatareanu M., 2009. Tough love: do Czech suppliers

learn from their relationships with multinationals? The Scandinavian

Journal of Economics 111 (4), 811-833.

Jovanovic B. and Braguinsky S., 2004. Bidder discounts and target premia

in takeovers. American Economic Review 94, 46-56.

Kaditi E. A. 2013. Foreign Investments and Institutional Convergence in

South-eastern Europe. International Economic Journal. 27, 109-126

Kaplan D., Piedra E. and Seira E., 2011. Entry regulation and business

start-ups: Evidence from Mexico. Journal of Public Economics 95, 1501-

1515.

Kindleberger C.P., 1969. American business abroad. The International

Executive 11, 11–12.

Kleinert J. and Toubal F., 2010. Gravity for FDI. Review of International

Economics 18, 1-13.

Knack S. and Keefer P., 1995. Institutions and economic performance:

Cross-country tests using alternative institutional measures. Economics

and Politics 7, 207-227.

Knickerbrocker F. T., 1973. Oligopolistic Reaction and Multinational

Enterprise. Cambridge, MA: Harvard University Press.

Kogut B. and Zander U., 1993. Knowledge of the firm and the evolutionary

theory of the multinational corporation. Journal of International Business

Studies 24, 625-645.

Krugman 1980. Scale economies, product differentiation, and the pattern of

trade. American Economic Review 70, 950-959.

225

Krugman P., 1991. Geography and Trade. The MIT Press.

LaFountain C., 2005. Where do firms locate? Testing competing models of

agglomeration. Journal of Urban Economics 58, 338-366.

Laursen K. and Masciarelli F., 2007. The Effect of regional Social Capital

and External Knowledge Acquisition on Process and Product Innovation.

ROCK Working Paper 043.

Ledayeva S., 2009. Spatial econometric analysis of foreign direct investment

determinants in Russian regions. The World Economy 32, 643-666.

Lee Y., 2003. Lean production systems, labour unions, and greenfield

locations of the Korean new auto assembly plants and their suppliers.

Economic Geography 79, 321-339.

Liu X., Lovely M. E. and Ondrich J., 2010. The location decisions of foreign

investors in China: Untangling the effect of wages using a control

function approach. The Review of Economics and Statistics 92, 160-166.

Liu X., Parker D., Vaidya K. and Wei Y., 2001. The impact of foreign direct

investment on labour productivity in the Chinese electronics industry.

International Business Review 10, 421-439.

Liu X., Siler P., Wang C. and Wei Y., 2000. Productivity spillovers from

foreign direct investment: Evidence from UK industry level panel data.

Journal of International Business Studies 31, 407-425.

Louri H., Papanastassiou M. and Lanoutirs J., 2000. FDI in the EU

periphery: A multinomial logit analysis of Greek firm strategies. Regional

Studies 34, 419-427.

Luo Y. and Tung R. L., 2007. International expansion of emerging market

enterprises: A springboard perspective. Journal of International Business

Studies 38, 481-498.

226

Maffini G., and Mokkas S., 2011. Profit sharing and measured productivity

of multinational firms. Oxford Bulletin of Economic and Statistics 73, 1-

19.

Marcin K., 2008. How does FDI inflow affect productivity of domestic firms?

The role of horizontal and vertical spillovers, absorptive capacity and

competition. The Journal of International Trade & Economic Development

17 (1), 155-173.

Mariotti S., Piscitello L. and Elia S., 2014. Local externalities and ownership

choices in foreign acquisitions by multinational enterprises. Economic

Geography, forthcoming.

Markusen J.R., 2001. Contracts, intellectual property rights, and

multinational investment in developing countries. Journal of International

Economics 53, 189-204.

Markusen J.R. and Venables A.J., 1999. Foreign direct investment as a

catalyst for industrial development. European Economic Review 43, 335-

356.

Mathews J.A., 2002. Competitive advantage of the latecomer firm: a

resource-based account of industrial catch-up strategies. Asia Pacific

Journal of Management 19, 467-488.

Mayer T, Mejean I. and Nefussi B., 2010. The location of domestic and

foreign production affiliates by French multinational firms. Journal of

Urban Economics 68, 115-128.

McCann P. and Acs Z.J., 2009. Globalisation: Countries, cities and

multinationals. Jena Economic Research Papers 042.

McCann P. and Mudambi R., 2005. Analytical differences in the economics

of geography: The case of the multinational firm. Environment and

Planning A 37, 1857-1876.

227

McFadden D., 1974. Conditional logit analysis of qualitative choice behavior.

In Zarembka P. (ed.), Frontiers in Econometrics. Academic Press, New

York.

Melitz M., 2003. The Impact of Trade on Intra-Industry Reallocations and